CN-121985313-A - Large pump station monitoring system, method, equipment and medium based on edge calculation

CN121985313ACN 121985313 ACN121985313 ACN 121985313ACN-121985313-A

Abstract

The invention provides a large pump station monitoring system, a method, equipment and a medium based on edge calculation, which relate to the technical field of intelligent monitoring, and are characterized in that operation state data of a pump set are collected through a plurality of sensors, a communication optimizing device is utilized to acquire optimized communication parameters by solving through a multi-objective optimizing model based on the current channel state so as to configure the communication parameters of the corresponding sensors, the operation state data are subjected to localized time sequence data cleaning and compression processing in the edge calculating device to acquire sensing data, and a cloud digital twin platform is used for carrying out visual display and state analysis based on a digital twin model of the sensing data-driven pump set, so that accurate mapping and intelligent analysis of the pump set state are realized, and monitoring instantaneity, system stability and intelligent operation and maintenance efficiency of the large pump set in a complex industrial environment are improved.

Inventors

ZENG XIUYING
CHEN JIAN
CHEN ZHIYUN
LI CHEN
Liao Ertai
CHEN WEI
QIU HAIYUN
LIANG JUWEN
GAO LIN
HUANG WENLONG
CHEN HONGSHENG
ZHENG YONG
XIA YONGLI
HE JIANKUN
XIONG TINGTING

Assignees

中水珠江规划勘测设计有限公司

Dates

Publication Date: 20260505
Application Date: 20260226

Claims (10)

1. Large-scale pump station monitoring system based on edge calculation, characterized by comprising: the monitoring device is used for collecting the running state data of the pump group through a plurality of sensors arranged on key nodes of the pump group; The communication optimizing device is in communication connection with the monitoring device and is used for acquiring the current channel state of the wireless network accessed by each sensor in the monitoring device, solving the current channel state to obtain optimized communication parameters by adopting a multi-objective optimizing model based on the current channel state, and sending the optimized communication parameters to the monitoring device so as to configure the communication parameters corresponding to the sensors, wherein the multi-objective optimizing model is an improved genetic algorithm taking the total network energy consumption and the network data packet collision probability as combined optimizing targets; the edge computing device is in communication connection with the monitoring device and is used for receiving the running state data sent by the sensor, and carrying out localized time sequence data cleaning and compression processing on the running state data to obtain sensing data; The cloud digital twin platform is in communication connection with the edge computing device and is used for receiving the perception data and driving the digital twin model of the pump set to perform visual display and state analysis based on the perception data.
2. The edge computation based large pump station monitoring system of claim 1, wherein the communication optimization means is further adapted to adaptively adjust the crossover probability and the variation probability in the improved genetic algorithm based on individual fitness and average fitness in the population in the genetic algorithm.
3. An edge computation based large pump station monitoring system according to claim 2, wherein the communication optimization means is further adapted to determine a total network energy consumption based on the transmission parameters and the transmission time of each of the sensors; Determining the collision probability of the network data packet based on the transmission time and the arrival rate of the data packet; and constructing an objective function of the joint optimization objective based on the total network energy consumption, the network data packet collision probability and a preset weight coefficient.
4. The large pump station monitoring system based on edge calculation according to claim 1, wherein the edge calculation device is further configured to perform outlier recognition and cleaning on the operation state data by adopting a sliding window weighted filtering algorithm, so as to obtain cleaned operation state data.
5. The edge computing-based large pump station monitoring system of claim 4, wherein the edge computing device is further configured to perform data compression on the cleaned operation state data based on a preset dead zone threshold and a maximum heartbeat interval to obtain compressed operation state data.
6. The edge computing-based large pump station monitoring system of claim 5, wherein the edge computing device is further configured to perform weighted fusion on vibration signal features and electrical signal features based on the cleaned operation state data and the compressed operation state data, and generate a comprehensive feature vector for the pump set in which the operation state is a health state.
7. The large pump station monitoring system based on edge calculation according to claim 6, wherein the edge calculation device is further configured to compare the integrated feature vector with a preset fault threshold, and if it is determined that the operation state of the pump set is an abnormal state, generate early warning information and send the early warning information to the cloud digital twin platform.
8. An edge computing-based large pump station monitoring method, characterized by being applied to an edge computing device of the edge computing-based large pump station monitoring system according to any one of claims 1 to 7, comprising: receiving pump set operation state data sent by a sensor; carrying out localized time sequence data cleaning and compression processing on the running state data to obtain sensing data; and sending the perception data to a cloud digital twin platform so as to drive a digital twin model of the pump set to perform visual display and state analysis.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the edge calculation based large pump station monitoring method of claim 8 when the computer program is executed.
10. A computer readable storage medium having stored thereon computer instructions which when executed by a processor implement a large pump station monitoring method according to claim 8 based on edge calculations.

Description

Large pump station monitoring system, method, equipment and medium based on edge calculation Technical Field The invention relates to the technical field of intelligent monitoring, in particular to a large pump station monitoring system, method, equipment and medium based on edge calculation. Background In operation and maintenance management of a large-scale water conservancy junction pump station, real-time accurate perception and state mapping of the whole system environment of the ultra-large power pump set are important, and safety, stability and scheduling efficiency of a cross-river basin water diversion project are directly related. In the prior art, the monitoring of a large-scale water conservancy junction pump station mainly depends on the laying of wired sensors or manual regular inspection, but because the internal space structure of the pump station is complex, metal equipment is dense and strong electromagnetic interference exists, the wired deployment construction is extremely difficult and the maintenance cost is high, the manual inspection is difficult to cover the state of the full period, and the hidden trouble of the fault is easy to leak or misjudge. In addition, although the conventional wireless sensing network (such as LoRa) can solve the wiring problem through wireless transmission, the direct application of standard communication parameters is difficult to adapt to the environment with serious multipath effect and large node concurrency in a pump station, channel conflict and data packet loss are extremely easy to cause, and the wireless sensing network is limited by the service life of a battery and is difficult to operate for a long time. Disclosure of Invention In view of the above, the present invention aims to provide a large pump station monitoring system, method, device and medium based on edge calculation, so as to improve reliability, instantaneity and intelligence level of pump station state monitoring in a complex industrial environment. In a first aspect, the present application provides a large pump station monitoring system based on edge computation, comprising: the monitoring device is used for collecting the operation state data of the pump group through a plurality of sensors arranged on key nodes of the pump group; the communication optimizing device is in communication connection with the monitoring device and is used for acquiring the current channel state of the wireless network accessed by each sensor in the monitoring device, solving the current channel state to obtain optimized communication parameters by adopting a multi-objective optimizing model based on the current channel state, and sending the optimized communication parameters to the monitoring device so as to configure the communication parameters of the corresponding sensors, wherein the multi-objective optimizing model is an improved genetic algorithm taking the total network energy consumption and the collision probability of network data packets as combined optimizing targets; The edge computing device is in communication connection with the monitoring device and is used for receiving the running state data sent by the sensor, and carrying out localized time sequence data cleaning and compression processing on the running state data to obtain sensing data; The cloud digital twin platform is in communication connection with the edge computing device and is used for receiving the perception data and driving a digital twin model of the pump set to perform visual display and state analysis based on the perception data. Optionally, the communication optimization device is further configured to adaptively adjust the crossover probability and the mutation probability in the improved genetic algorithm based on the fitness of individuals in the population in the genetic algorithm and the average fitness. Optionally, the communication optimizing device is further used for determining the total network energy consumption based on the transmission parameters and the transmission time of each sensor; determining the collision probability of the network data packet based on the transmission time and the arrival rate of the data packet; and constructing an objective function of the joint optimization objective based on the total network energy consumption, the collision probability of the network data packet and the preset weight coefficient. Optionally, the edge computing device is further configured to perform outlier recognition and cleaning on the running state data by using a sliding window weighted filtering algorithm, so as to obtain cleaned running state data. Optionally, the edge computing device is further configured to perform data compression on the cleaned running state data based on a preset dead zone threshold and a maximum heartbeat interval, so as to obtain compressed running state data. Optionally, the edge computing device is further configured to perform weighted fusion on the vibration sig