CN-121979037-A - Intelligent control system and method for deep water hydraulic automatic hooking beam

CN121979037ACN 121979037 ACN121979037 ACN 121979037ACN-121979037-A

Abstract

An intelligent control system and method for a deep water hydraulic automatic hooking beam belong to the technical field of deep water hydraulic control. The system comprises a multi-sensor fusion module, an AI control unit, a redundant hydraulic circuit and a remote monitoring module. The multi-sensor fusion module is used for collecting bolt displacement, a hydraulic system state and watertight cable health data in real time, the AI control unit comprises a fuzzy PID controller and a reinforcement learning optimization module, the control parameters are dynamically adjusted according to environmental disturbance, the redundant hydraulic circuit is automatically switched when a main circuit fails, and the watertight cable health detection adopts an impedance-high frequency signal dual-mode detection method. The control method comprises the steps of system initialization, dynamic environment modeling, fuzzy PID and reinforcement learning cooperative control, fault intelligent diagnosis, emergency response and the like. The hydraulic bolt positioning device solves the problems of low positioning precision and poor reliability of the hydraulic bolt in a deepwater environment, and remarkably improves the operation reliability and the intelligent level of a gate system of a hydropower station.

Inventors

YAO YONGHUAN
GAN HUIMIN
LI YAJUN
Deng lun
ZOU GAOQING
TONG YING
WU LONG

Assignees

中国长江电力股份有限公司

Dates

Publication Date: 20260505
Application Date: 20260114

Claims (13)

1. An intelligent control system of automatic couple roof beam of deep water hydraulic pressure, its characterized in that includes: the multi-sensor fusion module is used for collecting plug pin displacement, hydraulic system state and watertight cable health data in real time; The AI control unit is connected with the multi-sensor fusion module and comprises a fuzzy PID controller and a reinforcement learning optimization module, and is used for dynamically adjusting control parameters according to environmental disturbance; The redundant hydraulic circuit comprises a main hydraulic circuit and a standby hydraulic circuit, is connected with the AI control unit and is used for automatically switching when the main hydraulic circuit fails; The remote monitoring module is in communication connection with the AI control unit and is used for data transmission and remote instruction issuing; The multi-sensor fusion module comprises a high-precision displacement sensor, a watertight cable health detection unit and a hydraulic system monitoring sensor.
2. The intelligent control system of a deep water hydraulic automatic hooking beam according to claim 1, wherein the watertight cable health detection unit is used for executing an impedance-high frequency signal dual-mode detection method, comprising: The impedance measuring module is used for monitoring the cable impedance value in real time and comparing the cable impedance value with a reference value; The high-frequency signal injection module is used for injecting a 1MHz high-frequency signal into the cable and detecting the reflection coefficient; And the cable state evaluation module is used for comprehensively judging the cable insulation state according to the impedance change rate and the reflection coefficient.
3. The intelligent control system of a deep water hydraulic automatic hooking beam according to claim 1 or 2, characterized in that the AI control unit comprises: The dynamic environment modeling module is used for constructing a water flow disturbance prediction model based on the LSTM network; A fuzzy PID controller for dynamically adjusting PID parameters according to the displacement error and the error change rate; And the reinforcement learning optimization module is used for constructing a reward function based on the bolt positioning precision, the action time and the hydraulic energy consumption and iteratively optimizing the fuzzy rule base.
4. The intelligent control system of a deep water hydraulic autohook beam of claim 3, further comprising a fault diagnosis and emergency response module comprising: the LSTM fault diagnosis model is used for analyzing the multi-source time sequence sensor data and identifying the fault type and occurrence probability of the hydraulic system; the fault level dividing unit divides the faults into a level I, a level II and a level III; And the emergency strategy execution unit automatically executes corresponding emergency strategies according to the fault level, including pressure compensation, redundant loop switching or system emergency pressure relief.
5. The intelligent control method of the intelligent control system of the deep water hydraulic automatic hooking beam according to any one of claims 1 to 4, characterized in that it comprises the following steps: Initializing a system and calibrating parameters, and establishing sensor reference data and control parameters; dynamic environment modeling and data real-time acquisition are carried out, and water flow speed, water pressure and equipment state data are obtained; Adopting a fuzzy PID and reinforcement learning cooperative control algorithm to dynamically compensate the influence of environmental disturbance on the positioning of the latch; Performing intelligent fault diagnosis through an LSTM network, and executing automatic emergency response according to the fault level; And the data storage and remote collaborative optimization provide a performance optimization basis for long-term operation of the system.
6. The intelligent control method of the intelligent control system of the deep water hydraulic automatic hooking beam of claim 5, wherein the fuzzy PID and reinforcement learning cooperative control algorithm specifically comprises: Blurring bolt displacement errors and error change rates into a plurality of fuzzy subsets; calculating PID parameter correction values according to a preset fuzzy rule base; defining a state space, an action space and a reward function of reinforcement learning; and iteratively updating the fuzzy rule weight through a Q-learning algorithm, and optimizing the long-term control performance.
7. The intelligent control method of the intelligent control system of the deepwater hydraulic automatic hooking beam according to claim 5 or 6, wherein the watertight cable state monitoring adopts an impedance-high frequency signal dual-mode detection method, and the intelligent control method comprises the following steps: calculating the insulation attenuation rate of the cable, namely, delta K_ins= (Z_reference-Z (t))/Z_reference; Injecting a 1MHz high-frequency signal, and detecting a signal reflection coefficient gamma; when Δk_ins >0.2 or γ >0.1, it is determined that the cable is at risk of failure and an early warning is triggered.
8. The intelligent control method of the intelligent control system of the deep water hydraulic automatic hooking beam according to claim 7, wherein the fault intelligent diagnosis and emergency response comprises: normalizing and sliding window processing are carried out on pressure, flow and impedance time sequence data; Extracting time sequence features through a 3-layer LSTM network, and outputting fault categories and probabilities; When the I level fault is judged, performing pressure compensation and reinforcement monitoring; when the II-stage fault is judged, switching to a standby hydraulic circuit and performing deceleration operation; and when the III level fault is judged, the forced retraction of the bolt, the pressure relief of the system and the emergency alarm are executed.
9. The intelligent control method of the intelligent control system of the deep water hydraulic automatic hooking beam according to claim 8, wherein the LSTM fault diagnosis model adopts a sliding time window with 30 sampling points, inputs time sequence data including hydraulic pressure, flow and cable impedance of the past 30 seconds, and judges that the fault is effective when the confidence of the output fault category is more than or equal to 85 percent.
10. The intelligent control method of the intelligent control system of the deepwater hydraulic automatic hooking beam according to claim 9, wherein the dynamic environment modeling comprises: Establishing a water flow disturbance autoregressive moving average model ARMA (p, q), wherein p=3 and q=2; calculating the comprehensive water pressure P_total=ρgh+0.5.ρv2 according to the water depth h and the real-time movement speed v; acquiring actual position deviation of a gate slot through a visual recognition module, and constructing a space compensation matrix T; And integrating the water flow disturbance value, the water pressure variation and the gate slot position deviation into a comprehensive environment disturbance vector for compensating the control parameters in real time.
11. An intelligent control device, characterized in that an intelligent control method of an intelligent control system of a deep water hydraulic automatic hooking beam according to any one of claims 5-9 is adopted, comprising: the embedded processing unit is configured with a four-core ARM processor, and the main frequency is more than or equal to 1.2GHz and is used for running an AI control algorithm; The watertight junction box adopts a double-layer sealing structure, and a humidity sensor is arranged in the watertight junction box; the emergency power supply module is internally provided with a super capacitor group and can provide standby power for more than or equal to 30 minutes when the main power supply is interrupted; The underwater communication unit supports two modes of underwater acoustic communication and optical fiber communication; the anti-interference shielding structure adopts an electromagnetic shielding cover and a damping bracket, and meets the working environment requirement of 300 meters of water depth.
12. The intelligent control device of claim 11, wherein the software architecture of the embedded processing unit comprises: The real-time data acquisition layer has a sampling frequency of more than or equal to 200Hz and a data caching depth of more than or equal to 10 seconds; The environment modeling middle layer is used for realizing real-time calculation of a water flow disturbance model, a water pressure change model and a door slot offset compensation model; The intelligent control core layer comprises a fuzzy PID controller, a reinforcement learning optimizer and an LSTM fault diagnosis engine; The decision execution layer generates a hydraulic valve control signal and alarm information according to the control instruction and the fault diagnosis result; The layers are communicated by adopting a publish-subscribe message mechanism, and the processing delay of a key control instruction is less than 10ms.
13. An intelligent control device according to claim 11, wherein the redundant hydraulic circuit employs a dual pump dual valve architecture, the main circuit and the backup circuit sharing a hydraulic tank but having independent filtration and cooling systems, and wherein the pin displacement fluctuation during switching is <0.5mm.

Description

Intelligent control system and method for deep water hydraulic automatic hooking beam Technical Field The invention belongs to the technical field of deep water hydraulic automatic hooking beam control, and particularly relates to an intelligent control system and method for a deep water hydraulic automatic hooking beam. Background In a gate control system of a hydropower station, a hydraulic automatic hooking beam is used as key connecting equipment and plays a core role of opening and closing and locking the gate. Under the working condition of a deep water environment (50-300 m), the traditional hydraulic automatic hooking beam faces serious control precision challenges, namely, the hydraulic bolt positioning precision is obviously reduced due to the comprehensive influence of dynamic environment factors such as water flow disturbance, water pressure fluctuation, mechanical vibration and the like, so that the bolt is misplaced with a door slot, and equipment blocking, sealing failure and even structural damage are caused. The existing solution mostly adopts a static control strategy and single sensor feedback, can not adapt to the dynamic change characteristic of the deepwater environment, and the bolt positioning error is usually more than 2mm and is far higher than the accuracy requirement of +/-0.5 mm allowed by deepwater operation. The technical bottleneck severely restricts the reliable operation and intelligent upgrading of the hydropower station deep water gate system. Disclosure of Invention The invention aims to solve the technical problems of low positioning precision and poor reliability of a hydraulic bolt in a deepwater environment and remarkably improve the operation reliability and the intelligent level of a gate system of a hydropower station. In order to solve the technical problems, the invention adopts the following technical scheme: an intelligent control system of a deep water hydraulic automatic hooking beam, comprising: the multi-sensor fusion module is used for collecting plug pin displacement, hydraulic system state and watertight cable health data in real time; The AI control unit is connected with the multi-sensor fusion module and comprises a fuzzy PID controller and a reinforcement learning optimization module, and is used for dynamically adjusting control parameters according to environmental disturbance; The redundant hydraulic circuit comprises a main hydraulic circuit and a standby hydraulic circuit, is connected with the AI control unit and is used for automatically switching when the main hydraulic circuit fails; The remote monitoring module is in communication connection with the AI control unit and is used for data transmission and remote instruction issuing; The multi-sensor fusion module comprises a high-precision displacement sensor, a watertight cable health detection unit and a hydraulic system monitoring sensor. Preferably, the watertight cable health detection unit is configured to perform an impedance-high frequency signal dual mode detection method, including: The impedance measuring module is used for monitoring the cable impedance value in real time and comparing the cable impedance value with a reference value; The high-frequency signal injection module is used for injecting a 1MHz high-frequency signal into the cable and detecting the reflection coefficient; And the cable state evaluation module is used for comprehensively judging the cable insulation state according to the impedance change rate and the reflection coefficient. Preferably, the AI control unit includes: The dynamic environment modeling module is used for constructing a water flow disturbance prediction model based on the LSTM network; A fuzzy PID controller for dynamically adjusting PID parameters according to the displacement error and the error change rate; And the reinforcement learning optimization module is used for constructing a reward function based on the bolt positioning precision, the action time and the hydraulic energy consumption and iteratively optimizing the fuzzy rule base. Preferably, the system further comprises a fault diagnosis and emergency response module, wherein the fault diagnosis and emergency response module comprises: the LSTM fault diagnosis model is used for analyzing the multi-source time sequence sensor data and identifying the fault type and occurrence probability of the hydraulic system; the fault level dividing unit divides the faults into a level I, a level II and a level III; And the emergency strategy execution unit automatically executes corresponding emergency strategies according to the fault level, including pressure compensation, redundant loop switching or system emergency pressure relief. The intelligent control method of the intelligent control system of the deep water hydraulic automatic hooking beam comprises the following steps: Initializing a system and calibrating parameters, and establishing sensor reference data and control parameters; d