CN-116362713-B - Intelligent fishway operation and maintenance method and system based on machine learning

CN116362713BCN 116362713 BCN116362713 BCN 116362713BCN-116362713-B

Abstract

The invention discloses an intelligent fishway operation and maintenance method and an operation and maintenance system based on machine learning, wherein the method comprises the steps of obtaining coordinate information of monitoring points and corresponding water level monitoring information; the wireless network transmits water level monitoring information to the data receiving unit, the data operation unit calculates water level calculation information data corresponding to monitoring points of the fishway chamber, the mapping coefficients of the water level monitoring information and the water level calculation information data are determined according to different boundary conditions, the mapping coefficients are compared and matched, when the difference exceeds a preset threshold, the data operation unit sequentially determines positions of breakage or obstruction of each fishway chamber by adopting a machine learning algorithm, the coordinates of the monitoring points of the corresponding fishway chamber and the water level calculation information data are corrected, and the operation and maintenance prompting unit transmits the breakage condition and the obstruction condition of the fishway chamber to the monitoring terminal to carry out operation and maintenance treatment measures. The invention can realize the long-term sustainable operation of the fishway and exert the actual demand of the optimum ecological benefit of the fishway.

Inventors

DAI JIE
YAN TAO
YANG JI
MAO JINQIAO
Wang Jiapi
ZHANG JUNFANG
ZHANG WEIJU
Gong Dieqing
LU PENG

Assignees

河海大学

Dates

Publication Date: 20260512
Application Date: 20230220

Claims (8)

1. The intelligent fishway operation and maintenance method based on machine learning is characterized by comprising the following steps of: S1, acquiring coordinate information of monitoring points of a pool room and corresponding water level monitoring information by adopting a water level sensor, and defining the coordinate information and the corresponding water level monitoring information as a first set; s2, inputting boundary condition variables as initial conditions, using fishway and pool room basic parameters as constraint conditions, calculating the read monitoring point coordinate information as independent variables, calculating water level information data corresponding to the pool room monitoring points, and constructing the data and the corresponding pool room monitoring point coordinate information into a third set; S3, determining mapping coefficients of the second set and the third set according to different boundary conditions, and comparing and matching the data in the second set with the data in the third set; S4, determining the difference degree of the two sets according to the comparison and matching result in the step S3, and respectively performing the following actions according to the relation between the difference degree and a preset threshold value: (1) If the difference degree meets a preset threshold, no early warning is needed; (2) If the difference exceeds a preset threshold, screening abnormal element data in the second set and the third set to form a fourth set, dividing the data in the fourth set into a plurality of subsets according to the number of fishway cells, serving as a fifth set, and sequencing the fifth set; S5, adopting a machine learning algorithm to sequentially determine positions of the fishway pool chambers corresponding to each subset in the fifth set, which are damaged or blocked by obstacles, and correcting the coordinates of monitoring points and water level calculation information data of the corresponding pool chambers until the last subset is corrected; s6, transmitting the damage condition and the obstruction condition in the fishway pool room determined in the step S5 to a monitoring terminal for operation and maintenance treatment measures.
2. The machine learning-based intelligent fishway operation and maintenance method of claim 1, wherein the specific formula for calculating the water level information data of the pool room monitoring point in step S2 is as follows: ; Wherein, the For the ith cell, the coordinates of the monitoring point are as follows Water level information data at the location(s), To enter the flow in the ith cell chamber, Divided into a transverse length and a longitudinal length of the ith cell, Is that The average flow velocity of the transverse section where the monitoring point is located, Respectively is The lateral and longitudinal distances of the monitoring point from the i-th cell origin.
3. The machine learning based intelligent fishway operation and maintenance method of claim 2, wherein the water level information data is calculated to be the flow rate when entering the ith pool room The calculation method of (1) is as follows: ; ; Wherein, the As a flow coefficient of the water, the water is mixed with water, Subtracting the height of a single pool chamber sill from the water level upstream in the ith pool chamber, and g is the gravity acceleration; The fishway correction coefficient is respectively related to the fishway gradient and the pool indoor structure, The individual cell sill height is subtracted from the downstream water level in the ith cell.
4. The machine learning based intelligent fishway operation and maintenance method of claim 1, wherein the threshold is preset in step S4 Is determined by the following formula: ; Wherein, the The average flow of the inlet of the target fishway; The transverse length and the longitudinal length of the ith cell chamber are respectively; S is the gradient of a target fishway; The coefficient is corrected for the threshold value.
5. The machine learning-based intelligent fishway operation and maintenance method of claim 1, wherein the specific steps of step S5 are as follows: Extracting abnormal pool room monitoring point coordinate information in a first subset of the fifth subset, determining the position of the fishway pool room corresponding to the first subset where damage or obstruction occurs, and correcting the abnormal pool room monitoring point coordinate and water level monitoring data information in the first subset; calculating the coordinates of the monitoring point in the next pool room and the water level calculation information data by taking the elements in the corrected first subset as initial conditions, comparing and matching the coordinates with the second subset in the fifth set, and determining the position of the fishway pool room where damage or obstruction occurs and correcting the position; and so on until the last subset is corrected.
6. The machine learning-based intelligent fishway operation and maintenance method of claim 1, wherein in step S5, the condition of damage or obstruction in the fishway is determined by using a machine learning algorithm, and the specific formula is as follows: s501, determining geometric characteristics of the fishway; s502, aiming at different operation and maintenance conditions, modifying the water level information data calculation method Calculating the flow coefficient and water level information under different monitoring point coordinates; And S503, eliminating the data set with relatively small obstruction size and damage degree, and continuously training and verifying the rest data set.
7. The intelligent fishway operation and maintenance system based on machine learning is characterized by comprising a water level sensor module, a wireless network module, an energy supply module and a central control platform: The water level sensor module consists of a plurality of ultrasonic sensors and is used for accurately sensing water level information data of monitoring points in each pool chamber in the fishway; The wireless network module is used for receiving water level information data of monitoring points of each pool room in the fishway transmitted by the water level sensor module and uploading the monitoring point data and the corresponding water level information data to the central control platform; the energy supply module is used for supplying power to the water level sensor module, the wireless network module and the central control platform; the central control platform consists of a data receiving unit, a data processing unit, a data operation unit and an operation and maintenance prompting unit, wherein the data receiving unit is used for receiving pool room fishway water level information data transmitted by the water level sensor module through the wireless network module, the data processing unit is used for preprocessing the received data, the data operation unit is used for comparing the actual water level and the theoretical water level of the fishway, and the operation and maintenance prompting unit is used for making an operation and maintenance prompt according to an operation result; The data receiving unit is configured to acquire the coordinate information of the monitoring points of the pool room and the corresponding water level monitoring information, define the coordinate information as a first set, and define a plurality of different boundary conditions and the first set under the conditions as a second set; The data processing unit is configured to receive pool room fishway water level information data from the data receiving unit, calculate water level calculation information data corresponding to pool room monitoring points, define the water level calculation information data and corresponding pool room monitoring point coordinate information as a third set, determine mapping coefficients of the second set and the third set, compare and match the data in the second set with the data in the third set, determine the difference degree of the two sets according to a comparison and matching result, and respectively perform the following actions according to the relationship between the difference degree and a preset threshold value: (1) If the difference degree meets a preset threshold, no early warning is needed; (2) If the difference exceeds a preset threshold, screening abnormal element data in the second set and the third set to form a fourth set, dividing the data in the fourth set into a plurality of subsets according to the number of fishway cells, serving as a fifth set, and sequencing the fifth set; The data operation unit is configured to sequentially determine positions of the fishway pool chambers corresponding to each subset in the fifth set, which are damaged or blocked by the obstacle, by adopting a machine learning algorithm, and correct the coordinates of monitoring points and water level calculation information data of the corresponding pool chambers until the last subset is corrected; the operation and maintenance prompting unit is configured to transmit the damage condition and the obstruction condition in the fishway pool room determined in the data operation unit to the monitoring terminal for operation and maintenance treatment measures.
8. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method of any one of claims 1 to 6 when the computer program is executed by the processor.

Description

Intelligent fishway operation and maintenance method and system based on machine learning Technical Field The invention relates to the technical field of hydraulic engineering, in particular to an intelligent fishway operation and maintenance system and an operation and maintenance method thereof. Background The fishway is used as an effective way capable of effectively communicating upstream and downstream water flows of a gate and a dam and providing an upstream and downstream channel for fish, and is adopted as an important measure for fish protection in hydraulic and hydroelectric engineering construction. However, the technology of the domestic fishway engineering is not mature enough, and the newly built fishway rarely achieves the expected fish effect. In reality, many fishways lack regular maintenance and overhaul, the original ecological functions are lost, and various problems are exposed in the aspects of monitoring, management and maintenance of the fishways, such as sediment accumulation or obstruction in the fishways can influence the orifice of the bottom hole pool type fishway, the fishway fish passing efficiency is seriously influenced, even the fishway operation is failed, and for example, more than 1/3 fishway of a part of fishway water inlets is destroyed, a sand beam is formed at the fishway water outlets to isolate the fishway water inlets from a main river, and the fishway functions are lost. Therefore, a solution is needed to accurately judge the fault condition of the fishway, overcome the limitation of manual monitoring and the difficulty of remote operation and maintenance, and realize the real requirements of long-term sustainable operation of the fishway and exerting the optimal ecological benefit of the fishway. Disclosure of Invention The invention aims to solve the technical problem of providing the intelligent fishway operation and maintenance method based on machine learning, which can overcome the limitation of manual monitoring and the difficulty of remote operation and maintenance, and realize the real requirements of long-term sustainable operation of the fishway and the exertion of the optimal ecological benefit of the fishway. The invention adopts the following technical scheme for solving the technical problems: the intelligent fishway operation and maintenance method based on machine learning provided by the invention comprises the following steps: s1, acquiring coordinate information of monitoring points of a pool chamber and corresponding water level monitoring information by adopting a water level sensor, defining the coordinate information and the corresponding water level monitoring information as a first set, and defining a plurality of different boundary conditions and the first set under the conditions as a second set. S2, inputting boundary condition variables as initial conditions, using fishway and pool room basic parameters as constraint conditions, calculating the read monitoring point coordinate information as independent variables, calculating water level information data corresponding to the pool room monitoring points, and constructing the data and the corresponding pool room monitoring point coordinate information into a third set. When the upstream boundary condition changes, the third set is calculated and updated simultaneously. S3, determining mapping coefficients of the second set and the third set according to different boundary conditions, and comparing and matching the data in the second set with the data in the third set. The mapping coefficient is determined mainly by the fishway pool chamber and related operation conditions, if the fishway is large, the operation boundary condition is stable, the mapping coefficient can be biased to be loose, and if the fishway is small in size and sensitive to the change of the boundary condition, the mapping coefficient is severe. S4, determining the difference degree of the two sets according to the comparison and matching result in the step S3, and respectively performing the following actions according to the relation between the difference degree and a preset threshold value: (1) If the difference degree meets a preset threshold, no early warning is needed; (2) If the difference exceeds a preset threshold, abnormal element data in the second set and the third set are screened to form a fourth set, the data in the fourth set are divided into a plurality of subsets according to the number of fishway cells to be used as a fifth set, and the fifth set is ordered. S5, a machine learning algorithm is adopted, positions of the fishway pool chambers corresponding to each subset in the fifth set, which are damaged or blocked, are sequentially determined, and monitoring point coordinates and water level calculation information data of the corresponding pool chambers are corrected until the last subset is corrected. S6, transmitting the damage condition and the obstruction condition in the fishway pool room d