Multi-working-condition intelligent control system and method for hydraulic control butterfly valve of circulating water pump

Abstract

The invention relates to the technical field of fluid control, in particular to a multi-working-condition intelligent control system and method for a hydraulic control butterfly valve of a circulating water pump. According to the technical scheme, the multi-working-condition intelligent control system and method for the hydraulic control butterfly valve of the circulating water pump comprises a multi-working-condition self-adaptive control module, an intelligent decision control module, a fault diagnosis and fault tolerance module, a system integration and communication module and a real-time monitoring and prediction module. According to the invention, fundamental improvement is realized through the multi-working-condition self-adaptive control module, system operation data is acquired in real time by using a high-precision pressure sensor, a flow meter and a vibration sensor through an integrated working condition intelligent identification unit, millisecond-level working condition switching is realized through an edge computing node based on a working condition mode identification algorithm, and the system reliability is improved. Working condition modes such as a centrifugal pump, an axial flow pump and variable frequency speed regulation are accurately recognized, the adaptability of the system to multiple working conditions is improved, and the flexibility and energy efficiency performance of the system are remarkably enhanced.

Description

Technical Field

[0001] This invention relates to the field of fluid control technology, and in particular to a multi-condition intelligent control system and method for a circulating water pump hydraulic butterfly valve. Background Technology

[0002] In industrial circulating water systems, hydraulic butterfly valves are key control devices whose performance directly affects the operating efficiency and safety of the entire system. Traditional hydraulic butterfly valve control systems mainly use simple on / off control or proportional control, which cannot adapt to complex and ever-changing operating conditions.

[0003] Existing circulating water pump hydraulic butterfly valve control systems have significant limitations when switching between multiple operating conditions. These limitations mainly manifest as delayed operating condition identification, rigid parameter adjustment, and lack of energy efficiency coordination. Traditional systems often use fixed thresholds or manual switching modes, which cannot identify the dynamic changes of complex operating conditions such as centrifugal pumps, axial flow pumps, or variable frequency speed regulation in real time. This results in parameters such as valve opening speed and holding pressure not being able to adaptively adjust according to actual operating conditions, often leading to energy waste or unstable operation. In addition, traditional systems lack system-level energy efficiency optimization mechanisms. The linkage control between pump sets and valves is limited to the basic start-stop level, failing to form a full-process energy efficiency coordination management system. This makes it difficult to achieve energy-saving goals and directly limits the flexibility and efficiency of the system in different operating scenarios.

[0004] To address the aforementioned shortcomings, this solution achieves fundamental improvements through a multi-condition adaptive control module. By integrating an intelligent condition identification unit, it utilizes high-precision pressure sensors, flow meters, and vibration sensors to collect system operation data in real time. Edge computing nodes, based on condition mode recognition algorithms, enable millisecond-level condition switching, accurately identifying operating modes such as centrifugal pumps, axial flow pumps, and variable frequency speed control. A parameter dynamic optimization unit, combining a digital twin model and neural network algorithms, dynamically adjusts parameters such as valve opening speed and holding pressure according to real-time operating conditions, ensuring the system is always in optimal operating condition. The energy efficiency collaborative management unit, through a power monitoring module and energy efficiency analysis chip, links pump sets and valves to achieve system-level energy efficiency optimization, eliminating energy waste caused by fixed parameters in traditional systems, improving the system's adaptability to multiple operating conditions, and significantly enhancing the system's flexibility and energy efficiency. Summary of the Invention

[0005] To overcome the significant limitations of existing circulating water pump hydraulic butterfly valve control systems during multi-condition switching, which are mainly manifested in lagging condition identification, rigid parameter adjustment, and lack of energy efficiency coordination, traditional systems often use fixed thresholds or manual switching modes, which cannot identify the dynamic changes of complex conditions such as centrifugal pumps, axial flow pumps, or variable frequency speed regulation in real time. This results in parameters such as valve opening speed and holding pressure not being able to adaptively adjust according to actual conditions, often leading to energy waste or unstable operation. In addition, traditional systems lack system-level energy efficiency optimization mechanisms, and the linkage control between pump sets and valves is only at the basic start-stop level, failing to form a full-process energy efficiency coordination management, making it difficult to achieve energy-saving goals, and directly limiting the flexibility and efficiency of the system under different operating scenarios.

[0006] The technical solution of this invention is: a multi-condition intelligent control system for a circulating water pump hydraulic butterfly valve, comprising the following modules: Multi-condition adaptive control module: used to achieve millisecond-level switching of multiple operating modes and system-level energy saving through intelligent identification of operating conditions, dynamic optimization of parameters and coordinated energy efficiency management; Intelligent decision control module: used to overcome the limitations of traditional PLC control algorithms and perform advanced intelligent decision-making; Fault diagnosis and fault tolerance module: used to eliminate the problems of crude fault diagnosis and lack of self-recovery capability in traditional systems; System integration and communication module: used for efficient communication and data fusion through protocol conversion, data governance and edge-cloud collaboration; Real-time monitoring and prediction module: This module is used to enhance the monitoring capabilities throughout the entire lifecycle, and to perform proactive maintenance through multi-parameter monitoring, data analysis, and predictive early warning.

[0007] As a preferred option, the multi-condition adaptive control module includes: A11: Intelligent operating condition identification unit, including a high-precision pressure sensor, flow meter, and vibration sensor, for real-time identification of the operating conditions of centrifugal pumps / axial flow pumps and variable frequency speed control; A12: Parameter dynamic optimization unit, including a programmable hydraulic controller, an electric regulating valve, and a digital twin model module, is used to automatically adjust valve opening speed and pressure holding parameters based on operating conditions; A13: Energy efficiency collaborative management unit, including power monitoring module, energy efficiency analysis chip and 5G communication module, used to link pump group and valve for system-level energy efficiency optimization.

[0008] As a preferred option, the intelligent decision control module includes: A21: Adaptive control unit, including an adaptive controller, real-time database, and industrial-grade AI chip, used to dynamically adjust control strategies based on real-time data; A22: Fault prediction unit, including oil sensor, acoustic emission sensor and predictive maintenance platform, used to predict various potential faults; A23: Decision optimization unit, including industrial decision engine, digital twin sand table and AR-assisted decision system, used to generate control decisions based on multi-objective optimization algorithm.

[0009] As a preferred option, the fault diagnosis and fault tolerance module includes: A31: Real-time monitoring unit, including pressure pulse sensor, temperature field scanner and leak detector, for comprehensive monitoring of valve status parameters; A32: Intelligent diagnostic unit, including a fault diagnosis expert system, a signal analysis module, and an alarm classification controller, used to accurately locate the root cause of the fault and issue graded alarms; A33: Fault-tolerant control unit, including redundant hydraulic locks, emergency power modules and automatic switching valve groups, used to automatically switch redundant systems to ensure continuous operation in the event of a failure.

[0010] Preferably, the system integration and communication module includes: A41: Protocol conversion unit, including OPC UA gateway, MQTT protocol converter and industrial switch, used to unify communication protocols of multiple devices; A42: Data Governance Unit, including a metadata management platform, a data cleaning module, and lake warehouse integrated storage, is used to standardize data formats and improve data quality; A43: Edge-Cloud Collaboration Unit, including edge computing gateway, industrial cloud platform and 5G slicing network, is used for edge intelligence and cloud collaborative computing.

[0011] As a preferred option, the real-time monitoring and prediction module includes: A51: Multi-parameter monitoring unit, including a multi-parameter sensor array, wireless data acquisition unit, and real-time database, used to synchronously collect multi-dimensional operating parameters; A52: Data Analysis Unit, including a big data analysis platform, GPU computing module, and feature extraction algorithm library, used for real-time processing and analysis of massive monitoring data; A53: Predictive and Early Warning Unit, including a predictive maintenance engine, an early warning classification module, and an AR visualization platform, is used to predict potential faults and issue early warnings based on AI models.

[0012] A multi-condition intelligent control method for a circulating water pump hydraulic butterfly valve includes the following steps: S11: Achieves intelligent identification of multiple operating conditions through a high-precision sensor array and edge computing, dynamically optimizes various parameters by combining digital twins and neural networks, and links with a 5G energy efficiency module to save system energy; S12: Improve fault prediction accuracy based on LSTM timing prediction and expert system, and achieve fault self-recovery through adaptive PID control and redundant hydraulic lock design; S13: Enables edge-cloud collaboration through OPC UA protocol conversion and lake warehouse integrated data governance, and completes the entire process closed loop by combining multi-parameter wireless sensor network and GPU stream computing engine to support proactive maintenance throughout the entire lifecycle.

[0013] As a preferred method, the following steps are included when performing intelligent identification and dynamic parameter optimization for multiple operating conditions: S21: High-precision pressure sensors, flow meters, and vibration sensors are used to collect system pressure, flow, and vibration data in real time. Millisecond-level identification of centrifugal pump, axial flow pump, and variable frequency speed control conditions is performed through edge computing nodes based on operating condition mode recognition algorithms. The identification results directly trigger control strategy switching. S22: A virtual system model is constructed based on the digital twin model module. The system performance under different working conditions is simulated through real-time data synchronization. Combined with neural network algorithms, the valve opening speed and pressure holding parameters are dynamically optimized. The optimized parameters are transmitted to the programmable hydraulic controller for execution in real time through the 5G communication module. S23: Deploy a power monitoring module to monitor the energy consumption of pump units and valves in real time, combine it with an energy efficiency analysis chip to calculate system-level energy efficiency indicators, and adjust the opening of the electric regulating valve through a PID adaptive algorithm to achieve optimal system-level energy efficiency control; S24: Utilizes fiber optic grating sensing technology to monitor valve displacement in real time, combines a leak detector to capture minute leak signals, performs spectrum analysis through a signal analysis module, accurately locates the leak point, and triggers an alarm grading controller to execute tiered early warnings; S25: Unifies multi-device communication protocols through the OPC UA gateway, converts the protocols into standardized data formats, and combines them with the MQTT protocol converter to achieve data transmission between edge nodes and the cloud, with a data transmission latency of ≤10ms. S26: Deploy a metadata management platform to standardize and govern multiple metadata types, combine it with a data cleaning module to remove noisy data, and use a lake warehouse integrated storage system to store structured and unstructured data in a unified manner; S27: Periodically calibrate multiple sensors using dedicated calibration tools to check whether the range accuracy meets the ±0.5% requirement. The calibration data is uploaded to the central calibration database via an industrial switch to form a traceable calibration record.

[0014] Preferably, the intelligent decision-making and fault prediction process includes the following steps: S31: A fault prediction model is constructed based on the LSTM time-series prediction algorithm. Hydraulic oil status data is collected through oil sensors and acoustic emission sensors, and combined with an expert system to predict potential faults, providing early warning of major faults 72 hours in advance; S32: Deploys an adaptive controller and an industrial-grade AI chip, collects system operation data through a real-time database, and dynamically adjusts the control strategy using reinforcement learning algorithms to adaptively adjust multiple parameters; S33: The fault diagnosis expert system performs multi-dimensional analysis of real-time monitoring data, and combines wavelet analysis algorithm to extract the root causes of various faults, triggering the alarm hierarchical controller to execute a three-level alarm and generate a maintenance work order; S34: Construct redundant hydraulic locks and emergency power modules, automatically switching to the redundant system when the main system fails, and seamlessly transferring valve control through automatic valve group switching; S35: Deploys an industrial decision engine and digital twin sandbox, generates control decision schemes through multi-objective optimization algorithms, and visualizes the decision-making process using an AR-assisted decision-making system, supporting operators to adjust decision parameters in real time and verify decision effects; S36: Real-time data is preprocessed at the edge via an edge computing gateway, combined with big data analysis on an industrial cloud platform, and feature extraction and pattern recognition are performed using a GPU computing module; S37: Deploy a predictive maintenance engine and early warning classification module, build a predictive model based on digital twin + AI prediction technology, and display the prediction results and maintenance suggestions through an AR visualization platform to support maintenance personnel in formulating maintenance plans and performing maintenance operations based on the prediction results.

[0015] Preferably, the following steps are included when performing system integration and real-time monitoring and coordinated control: S41: Achieves unified conversion of communication protocols among multiple devices through a protocol conversion unit, deploys an OPC UA gateway and MQTT protocol converter to convert various protocols into standardized data formats, and combines with industrial switches for high-speed data transmission; S42: Deploy a metadata management platform and data cleaning module to standardize and govern various metadata, and combine it with lake warehouse integrated storage to achieve unified storage of structured and unstructured data. Improve data availability through AI-driven data quality control; S43: Build an edge computing gateway and industrial cloud platform, perform edge intelligence and cloud collaborative computing through 5G slicing network, deploy GPU computing modules and feature extraction algorithm library, and perform edge-side preprocessing and cloud-based deep analysis of real-time data; S44: Deploy a multi-parameter sensor array and wireless data acquisition unit to collect multi-dimensional operating parameters in real time, and store and quickly query the data through a real-time database; S45: Real-time analysis of monitoring data is performed through a big data analysis platform and feature extraction algorithm library. The stream computing engine is used for real-time data processing and pattern recognition. Combined with the predictive maintenance engine, maintenance suggestions are generated and the early warning classification module is triggered to execute classified early warnings. S46: Deploy an AR visualization platform and early warning classification module to display the system status in real time using AR technology, and combine the early warning classification module to provide classified early warnings for faults; S47: Build a system integration test platform to conduct integration tests on multiple units, verify the functions and performance indicators of the integrated system, and verify the stability and reliability of the system under high load conditions through stress testing.

[0016] The beneficial effects of this invention are: 1. Existing traditional circulating water pump hydraulic butterfly valve control systems have significant limitations when switching between multiple operating conditions. These limitations primarily manifest as delayed operating condition identification, rigid parameter adjustment, and a lack of energy efficiency coordination. Traditional systems often employ fixed thresholds or manual switching modes, failing to recognize the dynamic changes in complex operating conditions such as centrifugal pumps, axial flow pumps, or variable frequency speed control in real time. This results in parameters such as valve opening speed and holding pressure failing to adaptively adjust according to actual operating conditions, often leading to energy waste or operational instability. Furthermore, traditional systems lack a system-level energy efficiency optimization mechanism; the linkage control between pump units and valves is limited to basic start-stop levels, failing to achieve full-process energy efficiency coordination management and thus hindering the achievement of energy-saving goals. This directly limits the system's flexibility and efficiency under different operating scenarios. This solution addresses these limitations by implementing a multi-condition adaptive control module to achieve... This fundamental improvement integrates an intelligent operating condition identification unit, utilizing high-precision pressure sensors, flow meters, and vibration sensors to collect system operation data in real time. Edge computing nodes, based on operating condition mode recognition algorithms, enable millisecond-level operating condition switching, accurately identifying operating modes such as centrifugal pumps, axial flow pumps, and variable frequency speed control. A dynamic parameter optimization unit, combining a digital twin model and neural network algorithms, dynamically adjusts parameters such as valve opening speed and pressure holding based on real-time operating conditions, ensuring the system is always in optimal operating condition. The energy efficiency collaborative management unit, through a power monitoring module and energy efficiency analysis chip, links pump units and valves to achieve system-level energy efficiency optimization, eliminating energy waste caused by fixed parameters in traditional systems, improving the system's adaptability to multiple operating conditions, and significantly enhancing the system's flexibility and energy efficiency. 2. Existing traditional circulating water pump hydraulic butterfly valve control systems have significant shortcomings in fault diagnosis and fault tolerance. Fault diagnosis relies heavily on manual experience or basic sensor signals, lacking precise location capabilities, often resulting in misdiagnosis or missed diagnosis, leading to low maintenance efficiency. Furthermore, traditional systems lack self-recovery mechanisms; once a fault occurs, manual intervention or system shutdown for repair is required, severely impacting continuous system operation. This not only increases maintenance costs but also reduces system reliability and safety. This solution achieves a systemic improvement through a fault diagnosis and fault tolerance module, employing pressure pulse sensors, temperature field scanners, and leak detectors to monitor valve status parameters in all dimensions, combined with fiber optic... Displacement monitoring is achieved using grid sensing technology, accurately capturing minute leakage signals. The intelligent diagnostic unit integrates a fault diagnosis expert system and a signal analysis module, extracting fault characteristic frequencies through wavelet analysis algorithms to accurately locate the root causes of faults such as valve jamming and hydraulic lock failure, and triggering an alarm tiered controller to execute tiered early warnings. The fault-tolerant control unit constructs redundant hydraulic locks and emergency power modules, automatically switching to the redundant system when the main system fails. Seamless transfer of valve control is achieved through automatic valve group switching. This not only enables accurate fault location and rapid response, but also enhances the system's self-recovery capability through redundancy design, fundamentally improving the system's reliability and safety. Attached Figure Description

[0017] Figure 1 The diagram shown is a schematic flowchart of the multi-condition intelligent control system framework for a circulating water pump hydraulic butterfly valve according to the present invention. Figure 2 The diagram shown is a schematic of the multi-condition intelligent control method for a circulating water pump hydraulic butterfly valve according to the present invention, which includes intelligent identification of multi-conditions and dynamic optimization of parameters. Figure 3 The diagram shown is a schematic of the intelligent decision-making and fault prediction process of a multi-condition intelligent control method for a circulating water pump hydraulic butterfly valve according to the present invention. Figure 4 The diagram shown is a schematic of the system integration and real-time monitoring collaborative control process of a multi-condition intelligent control method for a circulating water pump hydraulic butterfly valve according to the present invention. Detailed Implementation

[0018] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0019] Please see Figure 1 This invention provides an embodiment: a multi-condition intelligent control system for a circulating water pump hydraulic butterfly valve, comprising the following modules: Multi-condition adaptive control module: used to achieve millisecond-level switching of multiple operating modes and system-level energy saving through intelligent identification of operating conditions, dynamic optimization of parameters and coordinated energy efficiency management; Intelligent decision control module: used to overcome the limitations of traditional PLC control algorithms and perform advanced intelligent decision-making; Fault diagnosis and fault tolerance module: used to eliminate the problems of crude fault diagnosis and lack of self-recovery capability in traditional systems; System integration and communication module: used for efficient communication and data fusion through protocol conversion, data governance and edge-cloud collaboration; Real-time monitoring and prediction module: This module is used to enhance the monitoring capabilities throughout the entire lifecycle, and to perform proactive maintenance through multi-parameter monitoring, data analysis, and predictive early warning.

[0020] As a preferred option, the multi-condition adaptive control module includes: A11: Intelligent operating condition identification unit, including a high-precision pressure sensor, flow meter, and vibration sensor, for real-time identification of the operating conditions of centrifugal pumps / axial flow pumps and variable frequency speed control; A12: Parameter dynamic optimization unit, including a programmable hydraulic controller, an electric regulating valve, and a digital twin model module, is used to automatically adjust valve opening speed and pressure holding parameters based on operating conditions; A13: Energy efficiency collaborative management unit, including power monitoring module, energy efficiency analysis chip and 5G communication module, used to link pump group and valve for system-level energy efficiency optimization.

[0021] As a preferred option, the intelligent decision control module includes: A21: Adaptive control unit, including an adaptive controller, real-time database, and industrial-grade AI chip, used to dynamically adjust control strategies based on real-time data; A22: Fault prediction unit, including oil sensor, acoustic emission sensor and predictive maintenance platform, used to predict various potential faults; A23: Decision optimization unit, including industrial decision engine, digital twin sand table and AR-assisted decision system, used to generate control decisions based on multi-objective optimization algorithm.

[0022] As a preferred option, the fault diagnosis and fault tolerance module includes: A31: Real-time monitoring unit, including pressure pulse sensor, temperature field scanner and leak detector, for comprehensive monitoring of valve status parameters; A32: Intelligent diagnostic unit, including a fault diagnosis expert system, a signal analysis module, and an alarm classification controller, used to accurately locate the root cause of the fault and issue graded alarms; A33: Fault-tolerant control unit, including redundant hydraulic locks, emergency power modules and automatic switching valve groups, used to automatically switch redundant systems to ensure continuous operation in the event of a failure.

[0023] Preferably, the system integration and communication module includes: A41: Protocol conversion unit, including OPC UA gateway, MQTT protocol converter and industrial switch, used to unify communication protocols of multiple devices; A42: Data Governance Unit, including a metadata management platform, a data cleaning module, and lake warehouse integrated storage, is used to standardize data formats and improve data quality; A43: Edge-Cloud Collaboration Unit, including edge computing gateway, industrial cloud platform and 5G slicing network, is used for edge intelligence and cloud collaborative computing.

[0024] As a preferred option, the real-time monitoring and prediction module includes: A51: Multi-parameter monitoring unit, including a multi-parameter sensor array, wireless data acquisition unit, and real-time database, used to synchronously collect multi-dimensional operating parameters; A52: Data Analysis Unit, including a big data analysis platform, GPU computing module, and feature extraction algorithm library, used for real-time processing and analysis of massive monitoring data; A53: Predictive and Early Warning Unit, including a predictive maintenance engine, an early warning classification module, and an AR visualization platform, is used to predict potential faults and issue early warnings based on AI models.

[0025] Please see Figure 2-4 In this embodiment, a multi-condition intelligent control method for a circulating water pump hydraulic butterfly valve includes the following steps: S11: Achieves intelligent identification of multiple operating conditions through a high-precision sensor array and edge computing, dynamically optimizes various parameters by combining digital twins and neural networks, and links with a 5G energy efficiency module to save system energy; S12: Improve fault prediction accuracy based on LSTM timing prediction and expert system, and achieve fault self-recovery through adaptive PID control and redundant hydraulic lock design; S13: Enables edge-cloud collaboration through OPC UA protocol conversion and lake warehouse integrated data governance, and completes the entire process closed loop by combining multi-parameter wireless sensor network and GPU stream computing engine to support proactive maintenance throughout the entire lifecycle.

[0026] As a preferred method, the following steps are included when performing intelligent identification and dynamic parameter optimization for multiple operating conditions: S21: High-precision pressure sensors, flow meters, and vibration sensors are used to collect system pressure, flow, and vibration data in real time. Millisecond-level identification of centrifugal pump, axial flow pump, and variable frequency speed control conditions is performed through edge computing nodes based on operating condition mode recognition algorithms. The identification results directly trigger control strategy switching. S22: A virtual system model is constructed based on the digital twin model module. The system performance under different working conditions is simulated through real-time data synchronization. Combined with neural network algorithms, the valve opening speed and pressure holding parameters are dynamically optimized. The optimized parameters are transmitted to the programmable hydraulic controller for execution in real time through the 5G communication module. S23: Deploy a power monitoring module to monitor the energy consumption of pump units and valves in real time, combine it with an energy efficiency analysis chip to calculate system-level energy efficiency indicators, and adjust the opening of the electric regulating valve through a PID adaptive algorithm to achieve optimal system-level energy efficiency control; S24: Utilizes fiber optic grating sensing technology to monitor valve displacement in real time, combines a leak detector to capture minute leak signals, performs spectrum analysis through a signal analysis module, accurately locates the leak point, and triggers an alarm grading controller to execute tiered early warnings; S25: Unifies multi-device communication protocols through the OPC UA gateway, converts the protocols into standardized data formats, and combines them with the MQTT protocol converter to achieve data transmission between edge nodes and the cloud, with a data transmission latency of ≤10ms. S26: Deploy a metadata management platform to standardize and govern multiple metadata types, combine it with a data cleaning module to remove noisy data, and use a lake warehouse integrated storage system to store structured and unstructured data in a unified manner; S27: Periodically calibrate multiple sensors using dedicated calibration tools to check whether the range accuracy meets the ±0.5% requirement. The calibration data is uploaded to the central calibration database via an industrial switch to form a traceable calibration record.

[0027] Preferably, the intelligent decision-making and fault prediction process includes the following steps: S31: A fault prediction model is constructed based on the LSTM time-series prediction algorithm. Hydraulic oil status data is collected through oil sensors and acoustic emission sensors, and combined with an expert system to predict potential faults, providing early warning of major faults 72 hours in advance; S32: Deploys an adaptive controller and an industrial-grade AI chip, collects system operation data through a real-time database, and dynamically adjusts the control strategy using reinforcement learning algorithms to adaptively adjust multiple parameters; S33: The fault diagnosis expert system performs multi-dimensional analysis of real-time monitoring data, and combines wavelet analysis algorithm to extract the root causes of various faults, triggering the alarm hierarchical controller to execute a three-level alarm and generate a maintenance work order; S34: Construct redundant hydraulic locks and emergency power modules, automatically switching to the redundant system when the main system fails, and seamlessly transferring valve control through automatic valve group switching; S35: Deploys an industrial decision engine and digital twin sandbox, generates control decision schemes through multi-objective optimization algorithms, and visualizes the decision-making process using an AR-assisted decision-making system, supporting operators to adjust decision parameters in real time and verify decision effects; S36: Real-time data is preprocessed at the edge via an edge computing gateway, combined with big data analysis on an industrial cloud platform, and feature extraction and pattern recognition are performed using a GPU computing module; S37: Deploy a predictive maintenance engine and early warning classification module, build a predictive model based on digital twin + AI prediction technology, and display the prediction results and maintenance suggestions through an AR visualization platform to support maintenance personnel in formulating maintenance plans and performing maintenance operations based on the prediction results.

[0028] Preferably, the following steps are included when performing system integration and real-time monitoring and coordinated control: S41: Achieves unified conversion of communication protocols among multiple devices through a protocol conversion unit, deploys an OPC UA gateway and MQTT protocol converter to convert various protocols into standardized data formats, and combines with industrial switches for high-speed data transmission; S42: Deploy a metadata management platform and data cleaning module to standardize and govern various metadata, and combine it with lake warehouse integrated storage to achieve unified storage of structured and unstructured data. Improve data availability through AI-driven data quality control; S43: Build an edge computing gateway and industrial cloud platform, perform edge intelligence and cloud collaborative computing through 5G slicing network, deploy GPU computing modules and feature extraction algorithm library, and perform edge-side preprocessing and cloud-based deep analysis of real-time data; S44: Deploy a multi-parameter sensor array and wireless data acquisition unit to collect multi-dimensional operating parameters in real time, and store and quickly query the data through a real-time database; S45: Real-time analysis of monitoring data is performed through a big data analysis platform and feature extraction algorithm library. The stream computing engine is used for real-time data processing and pattern recognition. Combined with the predictive maintenance engine, maintenance suggestions are generated and the early warning classification module is triggered to execute classified early warnings. S46: Deploy an AR visualization platform and early warning classification module to display the system status in real time using AR technology, and combine the early warning classification module to provide classified early warnings for faults; S47: Build a system integration test platform to conduct integration tests on multiple units, verify the functions and performance indicators of the integrated system, and verify the stability and reliability of the system under high load conditions through stress testing.

[0029] Example 1 Background: Existing circulating water pump hydraulic butterfly valve control systems suffer from deficiencies such as insufficient adaptability to multiple operating conditions, crude fault diagnosis, and low system integration, resulting in high energy consumption, high failure rate, and high maintenance costs. Taking a power plant as an example, the traditional system consumes 12 million kWh of energy annually, experiences an average of 15 shutdowns per year due to faults, and maintains 30% of the total equipment cost. Furthermore, switching between operating conditions requires manual intervention, with response times as long as several seconds, which cannot meet the requirements of smart power plants for high efficiency, reliability, and intelligence.

[0030] Implementation steps: S51: Install high-precision pressure sensors, flow meters, and vibration sensors at key locations on the pump body, valves, and pipelines to collect pressure, flow, and vibration data in real time; S52: Configure edge computing nodes to achieve millisecond-level identification of centrifugal pump / axial flow pump / variable frequency speed regulation operating conditions based on operating condition pattern recognition algorithms, triggering control strategy switching; S53: Dynamically optimizes valve opening speed and pressure holding parameters through a digital twin model and neural network algorithm, and transmits the data to the programmable hydraulic controller via a 5G communication module for execution; S54: Deploys a power monitoring module and energy efficiency analysis chip to link pump sets and valves, achieving system-level energy efficiency optimization and improving energy efficiency by 15%-20%; S55: Deploys adaptive controllers, real-time databases, and industrial-grade AI chips to dynamically adjust control strategies based on reinforcement learning algorithms; S56: Equipped with oil level sensors and acoustic emission sensors, and combined with LSTM time-series prediction algorithms, a fault prediction model is constructed to provide early warning of major faults up to 72 hours in advance, with a prediction accuracy of 92%. S57: Deploy an industrial decision engine, digital twin sandbox, and AR-assisted decision-making system to generate control decision schemes through multi-objective optimization algorithms, and combine them with an AR visualization platform to achieve real-time display and verification of the decision-making process; S58: Equipped with a pressure pulse sensor, temperature field scanner, and leak detector, combined with fiber optic grating sensing technology, it achieves 0.1mm-level displacement monitoring and captures valve status parameters in all dimensions; S59: Deploy a fault diagnosis expert system and signal analysis module to extract fault feature frequencies through wavelet analysis algorithms, accurately locate the root cause of the fault, and trigger a three-level alarm and generate a maintenance work order; S510: Constructs redundant hydraulic locks, emergency power modules, and automatic switching valve groups. In case of failure, it automatically switches to the redundant system, ensuring continuous operation time of ≥72 hours and improving system availability to 99.99%. S511: Deploy OPC UA gateway, MQTT protocol converter and industrial switch to unify the communication protocol of multiple devices and achieve data transmission latency of ≤10ms between edge nodes and the cloud; S512: Configures a metadata management platform, data cleaning module, and lake warehouse integrated storage to standardize data formats and improve data quality, increasing data availability to 98%; S513: Construct edge computing gateways, industrial cloud platforms, and 5G slicing networks to achieve edge intelligence and cloud-based collaborative computing, supporting a closed-loop process of real-time monitoring, analysis, and early warning; S514: Equipped with a multi-parameter sensor array, wireless data acquisition unit, and real-time database, it synchronously collects multi-dimensional operating parameters such as pressure, flow rate, vibration, and temperature; S515: Deploys a big data analytics platform, GPU computing module, and feature extraction algorithm library to process and analyze massive amounts of monitoring data in real time, and combines a predictive maintenance engine to generate maintenance suggestions and trigger tiered early warnings; S516: Integrates an AR visualization platform and early warning classification module, which uses AR technology to realize real-time visualization of system status, and supports maintenance personnel to formulate maintenance plans and perform maintenance operations based on the prediction results.

[0031] Data comparison table:

[0032] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A multi-condition intelligent control system for a circulating water pump hydraulic butterfly valve; characterized in that: It consists of the following modules: Multi-condition adaptive control module: used to achieve millisecond-level switching of multiple operating modes and system-level energy saving through intelligent identification of operating conditions, dynamic optimization of parameters and coordinated energy efficiency management; Intelligent decision control module: used to overcome the limitations of traditional PLC control algorithms and perform advanced intelligent decision-making; Fault diagnosis and fault tolerance module: used to eliminate the problems of crude fault diagnosis and lack of self-recovery capability in traditional systems; System integration and communication module: used for efficient communication and data fusion through protocol conversion, data governance and edge-cloud collaboration; Real-time monitoring and prediction module: This module is used to enhance the monitoring capabilities throughout the entire lifecycle, and to perform proactive maintenance through multi-parameter monitoring, data analysis, and predictive early warning.

2. The multi-condition intelligent control system for a circulating water pump hydraulic butterfly valve according to claim 1, characterized in that: The multi-condition adaptive control module includes: A11: Intelligent operating condition identification unit, including a high-precision pressure sensor, flow meter, and vibration sensor, for real-time identification of the operating conditions of centrifugal pumps / axial flow pumps and variable frequency speed control; A12: Parameter dynamic optimization unit, including a programmable hydraulic controller, an electric regulating valve, and a digital twin model module, is used to automatically adjust valve opening speed and pressure holding parameters based on operating conditions; A13: Energy efficiency collaborative management unit, including power monitoring module, energy efficiency analysis chip and 5G communication module, used to link pump group and valve for system-level energy efficiency optimization.

3. The multi-condition intelligent control system for a circulating water pump hydraulic butterfly valve according to claim 1, characterized in that: The intelligent decision-making and control module includes: A21: Adaptive control unit, including an adaptive controller, real-time database, and industrial-grade AI chip, used to dynamically adjust control strategies based on real-time data; A22: Fault prediction unit, including oil sensor, acoustic emission sensor and predictive maintenance platform, used to predict various potential faults; A23: Decision optimization unit, including industrial decision engine, digital twin sand table and AR-assisted decision system, used to generate control decisions based on multi-objective optimization algorithm.

4. The multi-condition intelligent control system for a circulating water pump hydraulic butterfly valve according to claim 1, characterized in that: The fault diagnosis and fault tolerance module includes: A31: Real-time monitoring unit, including pressure pulse sensor, temperature field scanner and leak detector, for comprehensive monitoring of valve status parameters; A32: Intelligent diagnostic unit, including a fault diagnosis expert system, a signal analysis module, and an alarm classification controller, used to accurately locate the root cause of the fault and issue graded alarms; A33: Fault-tolerant control unit, including redundant hydraulic locks, emergency power modules and automatic switching valve groups, used to automatically switch redundant systems to ensure continuous operation in the event of a failure.

5. The multi-condition intelligent control system for a circulating water pump hydraulic butterfly valve according to claim 1, characterized in that: The system integration and communication module includes: A41: Protocol conversion unit, including OPC UA gateway, MQTT protocol converter and industrial switch, used to unify communication protocols of multiple devices; A42: Data Governance Unit, including a metadata management platform, a data cleaning module, and lake warehouse integrated storage, is used to standardize data formats and improve data quality; A43: Edge-Cloud Collaboration Unit, including edge computing gateway, industrial cloud platform and 5G slicing network, is used for edge intelligence and cloud collaborative computing.

6. The multi-condition intelligent control system for a circulating water pump hydraulic butterfly valve according to claim 1, characterized in that: The real-time monitoring and prediction module includes: A51: Multi-parameter monitoring unit, including a multi-parameter sensor array, wireless data acquisition unit, and real-time database, used to synchronously collect multi-dimensional operating parameters; A52: Data Analysis Unit, including a big data analysis platform, GPU computing module, and feature extraction algorithm library, used for real-time processing and analysis of massive monitoring data; A53: Predictive and Early Warning Unit, including a predictive maintenance engine, an early warning classification module, and an AR visualization platform, is used to predict potential faults and issue early warnings based on AI models.

7. A multi-condition intelligent control system for a circulating water pump hydraulic butterfly valve according to any one of claims 1-6, characterized in that: A multi-condition intelligent control method for a circulating water pump hydraulic butterfly valve includes the following steps: S11: Achieves intelligent identification of multiple operating conditions through a high-precision sensor array and edge computing, dynamically optimizes various parameters by combining digital twins and neural networks, and links with a 5G energy efficiency module to save system energy; S12: Improve fault prediction accuracy based on LSTM timing prediction and expert system, and achieve fault self-recovery through adaptive PID control and redundant hydraulic lock design; S13: Enables edge-cloud collaboration through OPC UA protocol conversion and lake warehouse integrated data governance, and completes the entire process closed loop by combining multi-parameter wireless sensor network and GPU stream computing engine to support proactive maintenance throughout the entire lifecycle.

8. The multi-condition intelligent control method for a circulating water pump hydraulic butterfly valve according to claim 7, characterized in that: The following steps are included in the process of intelligent identification and dynamic parameter optimization under multiple operating conditions: S21: High-precision pressure sensors, flow meters, and vibration sensors are used to collect system pressure, flow, and vibration data in real time. Millisecond-level identification of centrifugal pump, axial flow pump, and variable frequency speed control conditions is performed through edge computing nodes based on operating condition mode recognition algorithms. The identification results directly trigger control strategy switching. S22: A virtual system model is constructed based on the digital twin model module. The system performance under different working conditions is simulated through real-time data synchronization. Combined with neural network algorithms, the valve opening speed and pressure holding parameters are dynamically optimized. The optimized parameters are transmitted to the programmable hydraulic controller for execution in real time through the 5G communication module. S23: Deploy a power monitoring module to monitor the energy consumption of pump units and valves in real time, combine it with an energy efficiency analysis chip to calculate system-level energy efficiency indicators, and adjust the opening of the electric regulating valve through a PID adaptive algorithm to achieve optimal system-level energy efficiency control; S24: Utilizes fiber optic grating sensing technology to monitor valve displacement in real time, combines a leak detector to capture minute leak signals, performs spectrum analysis through a signal analysis module, accurately locates the leak point, and triggers an alarm grading controller to execute tiered early warnings; S25: Unifies multi-device communication protocols through the OPC UA gateway, converts the protocols into standardized data formats, and combines them with the MQTT protocol converter to achieve data transmission between edge nodes and the cloud, with a data transmission latency of ≤10ms. S26: Deploy a metadata management platform to standardize and govern multiple metadata types, combine it with a data cleaning module to remove noisy data, and use a lake warehouse integrated storage system to store structured and unstructured data in a unified manner; S27: Periodically calibrate multiple sensors using dedicated calibration tools to check whether the range accuracy meets the ±0.5% requirement. The calibration data is uploaded to the central calibration database via an industrial switch to form a traceable calibration record.

9. The multi-condition intelligent control method for a circulating water pump hydraulic butterfly valve according to claim 7, characterized in that: The intelligent decision-making and fault prediction process includes the following steps: S31: A fault prediction model is constructed based on the LSTM time-series prediction algorithm. Hydraulic oil status data is collected through oil sensors and acoustic emission sensors, and combined with an expert system to predict potential faults, providing early warning of major faults 72 hours in advance; S32: Deploys an adaptive controller and an industrial-grade AI chip, collects system operation data through a real-time database, and dynamically adjusts the control strategy using reinforcement learning algorithms to adaptively adjust multiple parameters; S33: The fault diagnosis expert system performs multi-dimensional analysis of real-time monitoring data, and combines wavelet analysis algorithm to extract the root causes of various faults, triggering the alarm hierarchical controller to execute a three-level alarm and generate a maintenance work order; S34: Construct redundant hydraulic locks and emergency power modules, automatically switching to the redundant system when the main system fails, and seamlessly transferring valve control through automatic valve group switching; S35: Deploys an industrial decision engine and digital twin sandbox, generates control decision schemes through multi-objective optimization algorithms, and visualizes the decision-making process using an AR-assisted decision-making system, supporting operators to adjust decision parameters in real time and verify decision effects; S36: Real-time data is preprocessed at the edge via an edge computing gateway, combined with big data analysis on an industrial cloud platform, and feature extraction and pattern recognition are performed using a GPU computing module; S37: Deploy a predictive maintenance engine and early warning classification module, build a predictive model based on digital twin + AI prediction technology, and display the prediction results and maintenance suggestions through an AR visualization platform to support maintenance personnel in formulating maintenance plans and performing maintenance operations based on the prediction results.

10. The multi-condition intelligent control method for a circulating water pump hydraulic butterfly valve according to claim 7, characterized in that: When performing system integration and real-time monitoring and coordinated control, the following steps are included: S41: Achieves unified conversion of communication protocols among multiple devices through a protocol conversion unit, deploys an OPC UA gateway and MQTT protocol converter to convert various protocols into standardized data formats, and combines with industrial switches for high-speed data transmission; S42: Deploy a metadata management platform and data cleaning module to standardize and govern various metadata, and combine it with lake warehouse integrated storage to achieve unified storage of structured and unstructured data. Improve data availability through AI-driven data quality control; S43: Build an edge computing gateway and industrial cloud platform, perform edge intelligence and cloud collaborative computing through 5G slicing network, deploy GPU computing modules and feature extraction algorithm library, and perform edge-side preprocessing and cloud-based deep analysis of real-time data; S44: Deploy a multi-parameter sensor array and wireless data acquisition unit to collect multi-dimensional operating parameters in real time, and store and quickly query the data through a real-time database; S45: Real-time analysis of monitoring data is performed through a big data analysis platform and feature extraction algorithm library. The stream computing engine is used for real-time data processing and pattern recognition. Combined with the predictive maintenance engine, maintenance suggestions are generated and the early warning classification module is triggered to execute classified early warnings. S46: Deploy an AR visualization platform and early warning classification module to display the system status in real time using AR technology, and combine the early warning classification module to provide classified early warnings for faults; S47: Build a system integration test platform to conduct integration tests on multiple units, verify the functions and performance indicators of the integrated system, and verify the stability and reliability of the system under high load conditions through stress testing.

Images