Bionic fishway multi-parameter optimization method considering fishway environment and water quality demand

Abstract

This application relates to the field of fish conservation technology, specifically to a multi-parameter optimization method for biomimetic fishways that considers the environmental and water quality requirements of fishways. The method includes: obtaining the dissolved oxygen concentration in the fishway water; determining the guiding water temperature for guiding fish schools around high-flow-rate areas based on the difference between the dissolved oxygen concentration and a preset baseline oxygen demand; processing the positions of multiple fish school clusters to be guided and multiple adjustable guiding nodes based on a graph theory-based global optimization model to determine the optimal matching relationship between each fish school cluster and each adjustable guiding node; and controlling the corresponding adjustable guiding node to output guiding fluid with the guiding water temperature based on the optimal matching relationship. This application provides a method for coordinating the conflict between oxygenation operations and fish migration.

Description

Technical Field

[0001] This application relates to the field of fisheries management technology, specifically to a multi-parameter optimization method for biomimetic fishways that considers the environmental and water quality requirements of fishways. Background Technology

[0002] Fishways are core hydrological and ecological facilities that maintain the longitudinal connectivity of natural rivers and ensure the normal reproduction and survival of migratory fish. Through reasonable flow field design, they provide fish with a channel to swim upstream and are an important component of aquatic ecosystem restoration and protection. During the actual operation of fishways, various factors such as seasonal hydrological changes, fluctuations in upstream water quality, and changes in biological density within the fishway can lead to low dissolved oxygen concentrations in the main channel. Dissolved oxygen is a critical environmental factor for fish survival; oxygen-deficient water directly threatens fish lives. Therefore, oxygenation equipment is needed to oxygenate the water in the fishway.

[0003] To alleviate oxygen deficiency in fishway waters, current technologies typically involve installing aeration equipment within the main channel. When water quality sensors detect excessively low dissolved oxygen levels, the aeration equipment is activated to replenish the chemical oxygen content. However, the operation of this equipment creates a high-velocity turbulent flow zone downstream of its outlet. Since fish already exhibit reduced swimming ability under oxygen stress, this high-velocity turbulence exceeds their physiological resilience, preventing them from migrating upstream or even causing them to be swept away by the current. This creates a conflict between aeration operations and the flow rate requirements for fish migration. Therefore, how to coordinate the conflict between aeration operations and fish migration while maintaining adequate dissolved oxygen levels in the main channel is a pressing technical problem that needs to be solved. Summary of the Invention

[0004] To address the technical challenge of coordinating oxygenation operations with fish migration, this application aims to provide a multi-parameter optimization method for biomimetic fishways that considers both the fishway environment and water quality requirements. The specific technical solution adopted is as follows: Firstly, this paper provides a multi-parameter optimization method for an eco-friendly fishway that considers the environmental and water quality requirements of the fishway. This method includes: obtaining the dissolved oxygen concentration in the fishway water; determining the guiding water temperature for guiding fish to navigate high-velocity areas based on the difference between the dissolved oxygen concentration and a preset baseline oxygen demand; and using a graph theory-based global optimization model to process the positions of multiple fish clusters to be guided and multiple adjustable guiding nodes to determine the optimal matching relationship between each fish cluster and each adjustable guiding node. The optimization objective of the global optimization model is to minimize the comprehensive resistance cost overcome by all fish clusters as they move from their current positions to the matched guiding node. This comprehensive resistance cost is determined based on the spatial distance between the fish cluster and the guiding node and the local water flow velocity at the guiding node. Based on the optimal matching relationship, the corresponding adjustable guiding node is controlled to output a guiding fluid with the guiding water temperature.

[0005] In one possible design, a graph-based global optimization model processes the positions of multiple fish swarms to be guided and multiple adjustable guiding nodes to determine the optimal matching relationship between each fish swarm and each adjustable guiding node. This includes: constructing a first vertex set using multiple fish swarms to be guided and a second vertex set using multiple adjustable guiding nodes. The spatial distance between each fish swarm in the first vertex set and each guiding node in the second vertex set is calculated as a first environmental resistance parameter, and the local water flow velocity at each guiding node is obtained as a second environmental resistance parameter. The comprehensive resistance cost between each fish swarm and each guiding node is calculated based on the first and second environmental resistance parameters to construct a weighted bipartite graph. The maximum weighted matching in the weighted bipartite graph is solved to obtain the optimal matching relationship, where the objective of solving the maximum weighted matching corresponds to minimizing the sum of the comprehensive resistance costs of all matching pairs.

[0006] In one possible design, the comprehensive resistance cost between each fish swarm and each guiding node is calculated based on a first environmental resistance parameter and a second environmental resistance parameter. This includes: determining the maximum value among all first environmental resistance parameters as a first extreme value, and the maximum value among all second environmental resistance parameters as a second extreme value. A dimensionless distance characteristic value is determined based on the ratio of the spatial distance between each fish swarm and each guiding node to the first extreme value. A dimensionless velocity characteristic value is determined based on the ratio of the local water flow velocity at each guiding node to the second extreme value. The comprehensive resistance cost is determined by a weighted sum of the dimensionless distance characteristic value and the dimensionless velocity characteristic value.

[0007] In one possible design, before obtaining the dissolved oxygen concentration in the fishway water, the method further includes: obtaining a first dynamic parameter reflecting the operating status of the aeration equipment and a second dynamic parameter reflecting the spatial position of the fish school. When it is determined that the changing trends of the first dynamic parameter and the second dynamic parameter simultaneously meet a preset stagnation trigger condition, the second dynamic parameter at the current moment is determined as the swimming stagnation boundary distance, and the dissolved oxygen concentration and water temperature collected at the current moment are determined as the dissolved oxygen concentration and the baseline water temperature, respectively.

[0008] In one possible design, the first dynamic parameter is the output power of the aeration device, and the second dynamic parameter is the straight-line distance from the center of gravity of the fish school to a preset outlet reference point. The stagnation trigger condition is that the increase in output power is greater than zero and the increase in straight-line distance is non-negative.

[0009] In one possible design, a first vertex set is constructed using multiple fish clusters to be guided, including: obtaining the real-time positions of all fish clusters; comparing the real-time position of each fish cluster with the distance to a swimming stagnation boundary; extracting fish clusters whose real-time positions are outside the swimming stagnation boundary distance as the fish clusters to be guided, and constructing the first vertex set.

[0010] In one possible design, the guide water temperature for guiding fish to circumnavigate high-flow-rate areas is determined based on the difference between the dissolved oxygen concentration and the preset baseline oxygen demand. This includes: determining the dissolved oxygen concentration gap based on the difference between the dissolved oxygen concentration and the preset baseline oxygen demand; querying a preset metabolic compensation relationship table based on the dissolved oxygen concentration gap and the baseline water temperature to obtain the corresponding anoxic equivalent cooling value; and determining the guide water temperature based on the baseline water temperature and the anoxic equivalent cooling value.

[0011] In one possible design, when processing using a graph theory-based global optimization model, if the number of fish clusters to be guided is greater than the number of controllable guiding nodes, then for the remaining fish clusters that have not obtained independent matching relationships, collaborative guidance is provided through the collective behavior of the already matched fish clusters.

[0012] In one possible design, based on the optimal matching relationship, the corresponding adjustable guiding node is controlled to output guiding fluid with the guiding water temperature. This includes: determining the target guiding node to be activated based on the optimal matching relationship; controlling the target guiding node to activate its cooling and fluid output functions, allowing fluid at the guiding water temperature to seep into the edge of the fishway sidewall to create a temperature gradient field in the sidewall boundary layer region.

[0013] In one possible design, the method further includes: continuously acquiring the real-time dissolved oxygen concentration while controlling the output of the guide fluid from the adjustable guide node. When the real-time dissolved oxygen concentration recovers to above a preset baseline oxygen demand, or when the continuous output duration of the adjustable guide node reaches a preset safety threshold, the output of the guide fluid is stopped, and the power lock on the oxygenation equipment is released.

[0014] This application offers the following advantages: By using the dissolved oxygen concentration in the fishway water as a trigger condition and determining the guiding water temperature based on the dissolved oxygen gap, a graph theory-based global optimization model is used to optimally match multiple obstructed fish clusters with multiple adjustable guiding nodes without interfering with the normal operation of the main channel aeration equipment. This matching aims to minimize the overall resistance cost overcome by the fish migration, comprehensively considering the dual effects of spatial distance and local water flow velocity. Ultimately, the matched guiding nodes output low-temperature fluid, guiding the fish to bypass the high-velocity zone along the sidewall. This method transforms water quality parameters into induction signals and achieves optimal resource allocation between multiple fish clusters and multiple guiding nodes through global optimization. It effectively avoids secondary congestion caused by multiple fish clusters blindly gathering at a single node simultaneously. While ensuring the continuous operation of the main channel aeration, it provides obstructed fish with a low-energy, congestion-free parallel bypass channel, thereby coordinating the conflict between aeration operations and fish migration. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A flowchart illustrating a multi-parameter optimization method for an eco-friendly fishway that considers the fishway environment and water quality requirements, provided as an embodiment of this application. Figure 2 This is a schematic diagram of the structure of an eco-friendly fishway multi-parameter optimization device that takes into account the fishway environment and water quality requirements, provided in one embodiment of this application. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an eco-friendly fishway multi-parameter optimization method considering fishway environment and water quality requirements proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0019] The following description, in conjunction with the accompanying drawings, details a specific scheme for a multi-parameter optimization method for an eco-friendly fishway that considers both the fishway environment and water quality requirements, as provided in this application.

[0020] Please see Figure 1 It illustrates a flowchart of a multi-parameter optimization method for an eco-friendly fishway that considers the fishway environment and water quality requirements, according to an embodiment of this application. Figure 1 As shown, the method includes the following steps S101-S104.

[0021] S101. Obtain the dissolved oxygen concentration in the fishway water.

[0022] Dissolved oxygen concentration is an important water quality parameter characterizing the amount of dissolved oxygen in a body of water, usually measured in milligrams per liter (mg / L). In fishway environments, dissolved oxygen concentration directly affects the respiratory metabolism and swimming ability of fish, and is a key indicator for assessing the suitability of the fish's living environment.

[0023] In some embodiments, dissolved oxygen concentration data can be acquired in real time using dissolved oxygen sensors deployed in the fishway water. The dissolved oxygen sensor can employ optical or electrochemical methods for measurement, and its sampling frequency can be set according to the size of the fishway and the characteristics of fish activity, for example, once per minute. The acquired dissolved oxygen concentration data can be transmitted to a control center via wired or wireless communication for subsequent processing.

[0024] In some embodiments, to ensure data reliability, multiple dissolved oxygen sensors can be deployed at different cross-sectional locations of the fishway. For example, at least one dissolved oxygen sensor can be arranged at multiple locations, such as upstream and downstream of the outlet of the aeration equipment and in the middle of the fishway chamber. The collected dissolved oxygen concentration can be the average of the dissolved oxygen concentrations collected by multiple dissolved oxygen sensors to avoid the randomness of single-point sampling and ensure the accuracy of parameter acquisition.

[0025] S102. Determine the guiding water temperature for guiding fish to circumnavigate high-flow-rate areas based on the difference between dissolved oxygen concentration and preset baseline oxygen demand.

[0026] The preset baseline oxygen demand (BOD) is a dissolved oxygen concentration threshold pre-set based on the physiological characteristics of the target fish species. It represents the minimum dissolved oxygen level required for the normal physiological activities of this fish species. When the actual dissolved oxygen concentration is lower than the preset baseline BOD, it indicates insufficient oxygen supply in the water, and the fish may face hypoxic stress. For example, for common freshwater migratory fish, the preset baseline BOD can be set at 6.5 mg / L.

[0027] Water temperature guidance refers to the water temperature parameter used to regulate fish behavior. As poikilothermic animals, fish have a metabolic rate closely related to water temperature. Under conditions of insufficient dissolved oxygen, appropriately lowering the water temperature can slow down the metabolic rate of fish, thereby reducing their oxygen consumption per unit time and enabling them to have a stronger ability to adapt to limited dissolved oxygen conditions. Simultaneously, by establishing a specific temperature distribution, fish can be guided to choose suitable swimming paths, thus avoiding high-flow-rate areas.

[0028] In some embodiments, firstly, a dissolved oxygen concentration gap is determined based on the difference between the dissolved oxygen concentration and a preset baseline oxygen demand. The dissolved oxygen concentration gap is the difference between the preset baseline oxygen demand and the actual dissolved oxygen concentration. If the dissolved oxygen concentration gap is greater than zero, it indicates insufficient dissolved oxygen; the larger the gap, the more severe the oxygen deficiency. If the dissolved oxygen concentration gap is less than or equal to zero, it indicates sufficient dissolved oxygen, and no temperature compensation guidance is required.

[0029] Furthermore, when the dissolved oxygen deficit is greater than zero, a pre-defined metabolic compensation table is consulted based on the dissolved oxygen deficit and baseline water temperature to obtain the corresponding equivalent cooling value for hypoxia. The metabolic compensation table is a pre-established mapping table used to describe the equivalent temperature reduction required to maintain normal metabolic levels in fish under different combinations of dissolved oxygen deficit and baseline water temperature. This metabolic compensation table can be constructed based on fish physiological experimental data and reflects the compensatory effect of water temperature reduction on the oxygen consumption rate of fish.

[0030] For example, the metabolic compensation relationship table can take the form shown in Table 1.

[0031] Metabolic Compensation Relationship Table Table 1 The metabolic compensation relationship table is queried using dissolved oxygen deficiency and baseline water temperature as input indices, and outputs the corresponding hypoxia-equivalent cooling value. The hypoxia-equivalent cooling value represents the extent to which the water temperature needs to be lowered to offset the impact of insufficient dissolved oxygen on fish metabolism.

[0032] The metabolic compensation relationship table described above can be obtained as follows: First, the standard metabolic oxygen consumption rate curves of the target fish species under different water temperature gradients are measured in a constant-temperature water tank. Then, the equivalent water temperature reduction compensation value required to maintain the basic survival of this type of fish when the dissolved oxygen concentration in the environmental water decreases is calculated in reverse. Subsequently, the metabolic compensation relationship table is obtained based on the mapping relationship between dissolved oxygen deficit, current water temperature, and equivalent temperature reduction value. Finally, the obtained metabolic compensation relationship table is written into a read-only memory for subsequent querying.

[0033] Subsequently, the guide water temperature is determined based on the baseline water temperature and the equivalent cooling value due to oxygen deficiency.

[0034] For example, the guide water temperature is the difference between the base water temperature and the equivalent cooling value due to oxygen deficiency.

[0035] It should be noted that the guide water temperature should be ensured to be no lower than the target fish species' low-temperature tolerance limit to avoid causing cold stress damage to the fish. In some embodiments, a minimum temperature threshold can be set, and when the calculated guide water temperature is lower than this threshold, the guide water temperature is set as the minimum temperature threshold.

[0036] The minimum temperature threshold is determined based on the biological characteristics of the target fish species, typically taking the lower limit of the temperature at which the fish can move normally. For example, 16°C might be used for warm-water fish, while 6°C might be used for cold-water fish. For instance, if the fish to be guided in the current waterway are warm-water fish, then the minimum temperature threshold would be 16°C.

[0037] This step achieves adaptive water temperature regulation based on dissolved oxygen concentration gap, compensating for the impact of insufficient dissolved oxygen on fish metabolism by lowering the temperature, and guiding fish to choose suitable swimming paths by utilizing their temperature tropism.

[0038] S103. A graph theory-based global optimization model is used to process the positions of multiple fish swarms to be guided and the positions of multiple controllable guidance nodes to determine the optimal matching relationship between each fish swarm to be guided and each controllable guidance node.

[0039] In fishway monitoring, a fish school cluster refers to a group of fish that are spatially clustered and exhibit similar behavioral characteristics. The spatial distribution information of fish schools can be obtained using underwater cameras and / or acoustic detection equipment, and clustering algorithms (such as K-means or DBSCAN) can be used to group individuals with similar spatial locations into the same fish school cluster. Each fish school cluster has its centroid, which can be obtained by calculating the average spatial location of all individuals within the cluster.

[0040] Adjustable guidance nodes are environmental control devices deployed in fishways, featuring fluid output and temperature regulation functions. They can release fluids of a specific temperature into the surrounding water, thereby creating a temperature gradient field in a localized area. Adjustable guidance nodes can be embedded in the sidewalls or bottom of the fishway, and their number and layout are designed according to the fishway geometry and fish guidance requirements.

[0041] The global optimization model in graph theory is a pre-trained model. Specifically, the global optimization model in graph theory is an optimization algorithm based on graph theory mathematical theory, used to solve the optimal allocation scheme in many-to-many matching problems. In the embodiments of this application, the matching problem between the fish swarm to be guided and the controllable guiding node can be modeled as a bipartite graph matching problem, where the optimization objective is to minimize the comprehensive resistance cost overcome by all fish swarms as they move from their current positions to the matching guiding node.

[0042] The overall resistance cost is a quantitative indicator that assesses the environmental resistance a fish swarm needs to overcome to migrate from its current location to the target guidance node. This indicator comprehensively considers both spatial distance and water flow resistance: the greater the spatial distance, the more energy the fish swarm needs to expend to swim; the higher the local water flow velocity, the greater the fluid resistance the fish swarm needs to overcome. Therefore, the overall resistance cost is determined based on the spatial distance between the fish swarm and the guidance node and the local water flow velocity at the guidance node.

[0043] In some embodiments, a first vertex set is constructed using multiple fish swarms to be guided, and a second vertex set is constructed using multiple adjustable guiding nodes. The spatial distance between each fish swarm in the first vertex set and each guiding node in the second vertex set is calculated as a first environmental resistance parameter, and the local water flow velocity at each guiding node is obtained as a second environmental resistance parameter. The comprehensive resistance cost between each fish swarm and each guiding node is calculated based on the first and second environmental resistance parameters to construct a weighted bipartite graph. The maximum weight matching of the weighted bipartite graph is solved to obtain the optimal matching relationship, wherein the objective of solving the maximum weight matching corresponds to minimizing the sum of the comprehensive resistance costs of all matching pairs. This step is detailed in the following embodiments and will not be repeated here.

[0044] S104. Based on the optimal matching relationship, control the corresponding adjustable guide node to output guide fluid with guide water temperature.

[0045] In some embodiments, if an optimal matching relationship is obtained, the target boot node to be activated is determined based on the optimal matching relationship.

[0046] Specifically, in this embodiment, the optimal matching relationship provides a one-to-one correspondence between each fish swarm to be guided and each adjustable guidance node. Based on this matching relationship, all matched adjustable guidance nodes are determined, and these matched adjustable guidance nodes are the target guidance nodes that need to be activated. Unmatched adjustable guidance nodes remain in a closed state to save energy and avoid unnecessary interference to the fish swarm.

[0047] Furthermore, the control target guide node activates its cooling and fluid output functions, and leaks fluid at the guide water temperature to the edge of the fishway sidewall to build a temperature gradient field in the sidewall boundary layer region.

[0048] In this embodiment, the target guidance node may integrate a cooling module and a fluid delivery module. The cooling module may employ a semiconductor cooling chip or a compressor cooling device to cool the fluid to the target temperature; the fluid delivery module may employ a micro pump or a pressure-driven device to deliver the cooled fluid to the node outlet.

[0049] The control system activates the cooling function of the target guidance node, cooling the internal fluid to the guide water temperature; simultaneously, it activates the fluid output function, releasing the cooled fluid through seepage from the node outlet into the water at the edge of the fishway sidewall. Due to the relatively low water flow velocity in the sidewall boundary layer region, the temperature gradient remains relatively stable, forming an attractive low-temperature zone.

[0050] A temperature gradient field refers to a spatially unevenly distributed temperature field where the temperature gradually increases from the guide node outlet outwards. Fish exhibit positive tropism, moving towards suitable temperature environments. When they sense a temperature gradient, they move towards the lower-temperature region (i.e., near the guide node), thus achieving a guiding effect for their migration towards the guide node. Simultaneously, since guide nodes are typically deployed near the sidewalls of the fishway, the fish naturally avoid the high-velocity areas of the main channel as they move towards the lower-temperature region, achieving a protective guidance by bypassing the high-velocity area.

[0051] Understandably, this embodiment achieves precise node control and temperature gradient field construction based on optimal matching results, which can create an attractive low-temperature environment for fish in specific locations and guide the fish to migrate along a low-resistance path.

[0052] In some embodiments, during the process of controlling the output of guiding fluid by the adjustable guiding node, the real-time dissolved oxygen concentration is continuously acquired. When the real-time dissolved oxygen concentration recovers to above a preset baseline oxygen demand, or when the continuous output duration of the adjustable guiding node reaches a preset safety threshold, the output of guiding fluid is stopped, and the power lock of the oxygenation equipment is released.

[0053] Specifically, during the guide fluid output, dissolved oxygen concentration changes in the fishway water are continuously monitored using a dissolved oxygen sensor to obtain real-time dissolved oxygen concentration data. The monitoring frequency can be set according to the rate of environmental change, for example, once every 30 seconds. The termination conditions for guide fluid output include at least one of the following two termination conditions: The first termination condition is that the real-time dissolved oxygen concentration recovers to above the preset baseline oxygen demand. When this condition is met, it indicates that the water body has sufficient oxygen supply, the fish are no longer facing hypoxia stress, there is no need to continue metabolic compensation through cooling, and the outflow of guiding fluid can be stopped.

[0054] The second termination condition is that the continuous output duration of the adjustable guide node reaches a preset safety threshold. The preset safety threshold is an upper limit set to prevent prolonged low-temperature guidance from causing cold adaptation burden or ecological behavioral disturbance to fish; for example, it can be set to 30 minutes. When this condition is met, the guide fluid output will stop even if dissolved oxygen has not fully recovered, in order to ensure the long-term physiological health of the fish.

[0055] When any termination condition is met, perform the following operations: stop all adjustable guide nodes that are outputting guide fluid and shut down their cooling and fluid output functions; unlock the power of the oxygenation equipment, allowing the oxygenation equipment to make normal power adjustments as needed, and restore the normal oxygen supply management mode.

[0056] Through the aforementioned dynamic monitoring and termination control mechanism, this step achieves adaptive adjustment of the guidance strategy, which can both stop unnecessary intervention in a timely manner after dissolved oxygen recovers and prevent excessive guidance from causing negative impacts on fish through a safety threshold.

[0057] In one design, in order to obtain the optimal matching relationship, the above S103 includes: S1031-S1034.

[0058] S1031. Construct a first set of vertices using multiple fish swarms to be guided, and construct a second set of vertices using multiple controllable guiding nodes.

[0059] When constructing the first vertex set, the real-time spatial distribution information of all fish clusters within the fishway is first obtained. Specifically, this is achieved by fusing the acoustic echo signals acquired by the side-scan sonar module with the video stream data collected by the visual monitoring equipment. A preset data fusion algorithm is then applied to spatially align and weightedly fuse the shallow visual coordinates and deep-water sonar coordinates, thereby extracting the overall centroid 3D coordinates of each fish cluster within a pre-established 3D pool coordinate system. For any given fish cluster, its centroid coordinates are determined by a weighted average of the center points of each fish detection frame, ensuring that the extracted coordinates accurately reflect the spatial position of the fish cluster within the pool. The preset data fusion algorithm can be a Kalman filter algorithm, an extended Kalman filter algorithm, a deep learning-based data fusion algorithm, or a weighted average fusion algorithm.

[0060] Simultaneously, the fixed physical coordinates of all adjustable guide nodes are extracted to construct a second vertex set. These adjustable guide nodes are hardware devices pre-positioned on both sides of the fishway, possessing independent cooling and fluid infiltration functions. Each node is spatially calibrated during installation in the pool chamber, and its three-dimensional coordinates are stored as fixed constants in memory. While extracting the coordinates of each node, miniature water flow velocity sensors positioned next to each guide node are used to read the local water flow velocity at that node's location in real time. This velocity data is bound to the node coordinates and stored for subsequent calculation of the overall resistance cost.

[0061] Based on this, the total number of fish swarms to be guided and the total number of controllable guiding nodes are counted separately. The number of fish swarms to be guided is taken as... The number of adjustable boot nodes is For example, the system uses... The spatial coordinates of a fish cluster are used to construct the first vertex set. , represented as , where the first vertex set middle Characterizing the first The three-dimensional coordinates of the centroid of a fish cluster. The second vertex set is constructed using the fixed coordinates of each guiding node and its corresponding local velocity. , represented as , where the set of second vertices middle Characterizing the first The data pairs that bind the spatial coordinates of each guiding node to the local flow velocity.

[0062] In graph theory, vertices are the basic elements that make up a graph and are used to represent the objects of study. In this step, each fish swarm to be guided is abstracted as a vertex in the first vertex set, and each controllable guiding node is abstracted as a vertex in the second vertex set. The two vertex sets together constitute the vertex set of a bipartite graph.

[0063] S1032. Calculate the spatial distance between each fish cluster in the first vertex set and each guiding node in the second vertex set as the first environmental resistance parameter, and obtain the local water flow velocity at each guiding node as the second environmental resistance parameter.

[0064] In some embodiments, after the first vertex set and the second vertex set are constructed, the spatial distance between each fish swarm to be guided in the first vertex set and each adjustable guiding node in the second vertex set is calculated as a first environmental resistance parameter, and the local water flow velocity at each adjustable guiding node is obtained as a second environmental resistance parameter.

[0065] Specifically, taking the first vertex set Includes There are several fish groups waiting to be guided, among which the first... The three-dimensional coordinates of the centroid of the fish cluster are: Second vertex set Includes The number of adjustable guiding nodes, among which the first... The fixed spatial coordinates of the guiding nodes in the three-dimensional pool chamber coordinate system are: The standard Euclidean distance formula is used to calculate the first... The fish cluster and the first The straight-line distance between each guiding node The calculation formula is as follows: in, Respectively characterize the first The x-coordinate, y-coordinate, and y-coordinate of the centroid of a fish group in a three-dimensional pool room coordinate system; Respectively characterize the first The horizontal, vertical, and axial coordinates of an adjustable guide node in the three-dimensional pool chamber coordinate system; This represents the physical distance that the fish need to travel from the current location of the fish swarm to the guiding node. The magnitude of this distance directly reflects the path length cost that the fish swarm needs to overcome to move. The greater the distance, the more energy the fish swarm needs to consume to move, and the lower its probability of successful migration under hypoxic stress.

[0066] In addition, by deploying water flow velocity sensors at each adjustable guide node, the data is collected in real time. Local water flow velocity at the location of each guiding node This water flow velocity sensor employs either thermal or electromagnetic velocity measurement principles. The probe is installed on the main channel side, adjacent to the seepage outlet of the guide node. The measurement range is the average water flow velocity within an adjustable radius of approximately 0.3 meters around the guide node. (Local water flow velocity...) This represents the intensity of water flow resistance that a school of fish must overcome when swimming in the vicinity of this node. The greater the flow velocity, the greater the resistance that the school of fish must overcome when swimming against the current. For a school of fish in a state of hypoxia stress, the difficulty of successfully reaching this node is even greater.

[0067] S1033. Calculate the comprehensive resistance cost between each fish cluster and each guiding node based on the first environmental resistance parameter and the second environmental resistance parameter, so as to construct a weighted bipartite graph.

[0068] In some embodiments, the maximum value among all first environmental resistance parameters is determined as a first extreme value, and the maximum value among all second environmental resistance parameters is determined as a second extreme value; a dimensionless distance characteristic value is determined based on the ratio of the spatial distance between each fish swarm and each guiding node to the first extreme value; a dimensionless velocity characteristic value is determined based on the ratio of the local water flow velocity at each guiding node to the second extreme value. The overall resistance cost is determined based on the weighted sum of the dimensionless distance characteristic value and the dimensionless velocity characteristic value.

[0069] Specifically, the first and second environmental resistance parameters are first processed to be dimensionless to eliminate the influence of different physical dimensions on subsequent weighted calculations. Since the dimension of spatial distance is length (e.g., meter), the local water flow velocity... The units of measurement are velocity units (e.g., meters per second). Since these two parameters belong to different dimensions in a physical sense, they cannot be directly added or subtracted. Therefore, the maximum value among all first environmental resistance parameters is extracted as the first extreme value, and the maximum value among all second environmental resistance parameters is extracted as the second extreme value.

[0070] For example, the maximum spatial distance between all fish clusters and the guide node is taken as... The maximum local flow velocity at all guiding nodes is For example, based on the ratio of the spatial distance between each fish cluster and each guiding node to the first extreme value, the formula for determining the dimensionless distance feature value is as follows: in, For the first The fish cluster and the first The spatial straight-line distance between adjustable guiding nodes For the first The fish cluster and the first The dimensionless distance characteristic value corresponding to the linear spatial distance between adjustable guiding nodes.

[0071] Based on the ratio of the local flow velocity at each guiding node to the second extreme value, the formula for calculating the dimensionless flow velocity characteristic value is as follows: in, For the first The fish cluster and the first The spatial straight-line distance between adjustable guiding nodes Characterizing the first The fish cluster and the first The dimensionless eigenvalue of the spatial distance between each guiding node after normalization ranges from 0 to 1. The larger the value, the longer the spatial distance is relative to the global maximum distance. To prevent calculation overflow caused by a zero denominator, a very small fixed positive number is introduced. The resulting calculation error is within the allowable range. Characterizing the first The dimensionless characteristic value of the local flow velocity at each guiding node after normalization also ranges from 0 to 1. The larger the value, the stronger the velocity resistance at that node is relative to the global maximum velocity. To avoid calculation overflow caused by a zero denominator, a very small fixed positive number is introduced. The resulting calculation error is within the allowable range.

[0072] Through the above normalization process, spatial distance and local water flow velocity are uniformly mapped to the same numerical range, eliminating the difference in dimensions, and allowing the two to be weighted and combined under the same mathematical framework.

[0073] After dimensionless processing, preset distance and velocity penalty coefficients are introduced to weight and sum the dimensionless distance and velocity characteristic values ​​to obtain the comprehensive swimming resistance evaluation value. The sum of the distance and velocity penalty coefficients is 1, representing the influence weights of spatial distance and local flow velocity in the comprehensive resistance evaluation, respectively. A minimum constant is introduced to prevent the denominator from being zero in subsequent reciprocal calculations. The formula for calculating the comprehensive swimming resistance evaluation value is as follows: in, The preset distance penalty coefficient, This is the preset flow rate penalty coefficient.

[0074] In actual engineering deployment, and The value is preset through the system configuration file, and maintenance personnel can select the appropriate configuration scheme according to the actual working conditions of the fishway. For example, the default configuration adopts the balanced distribution mode of Example 1. , This configuration achieves a relatively balanced guiding effect under most operating conditions. When maintenance personnel find that the guiding effect under specific operating conditions needs optimization through monitoring data analysis, they can adjust the weight parameters through the human-machine interface. For example, when the output power of the oxygenation equipment increases or decreases, the weight coefficient adjustment is triggered. When the power increases, the system increases... The value of is determined with an emphasis on distance factors; when power is reduced, the system should appropriately reduce . The value of is determined by the weighting of distance and flow velocity.

[0075] Through the above-mentioned value selection rules and configuration schemes, α and β can not only reflect the differences in the physical influence of spatial distance and local flow velocity during the fish migration process, but also have the ability to be flexibly adjusted for different working conditions and fish species, providing a scientific and reasonable weight input for the global optimization model.

[0076] To apply the comprehensive resistance cost to the maximum weight matching algorithm, the comprehensive swimming resistance evaluation value is transformed into the inverse weight of swimming energy consumption through a reciprocal transformation. Since the objective of maximum weight matching is to maximize the sum of the weights of the matching edges, while the optimization objective of this application is to minimize the sum of the comprehensive resistance costs of all fish cluster transfers, the reciprocal operation is used to make the weight of edges with larger comprehensive swimming resistance evaluation values ​​smaller. Therefore, during the maximum weight matching solution process, the algorithm will automatically favor edges with larger weights, i.e., transfer paths with lower comprehensive swimming resistance. The formula for calculating the inverse weight of swimming energy consumption is: in, For the first The fish cluster and the first The inverse weight of the movement energy consumption of the matching edge between each guiding node is the reciprocal of the comprehensive movement resistance evaluation value, and the two are strictly inversely negatively correlated. It is a very small fixed number, and its function is to ensure that... When the denominator approaches zero, it is not zero to ensure the validity of the calculation result. In practical applications, it can be set to 0.001. The resulting calculation error is within the allowable range.

[0077] After calculating the inverse weights of swimming energy consumption between all fish swarms and all guiding nodes, the first set of vertices is used as the basis for determining the energy consumption. A cluster of fish to be guided is taken as one vertex of the bipartite graph, and the second vertex set is used as the basis for further analysis. An adjustable guiding node is used as the other vertex of the bipartite graph to calculate the weights. The weights of the edges connecting corresponding vertices are used to construct the complete weighted bipartite graph structure. If the number of vertices in the first vertex set and the second vertex set are not equal, virtual placeholder vertices are added to the side with fewer vertices before constructing the weighted bipartite graph to make the number of vertices on both sides equal. Edges involving virtual vertices are assigned a weight of zero to ensure that subsequent graph theory algorithms can handle non-matrix inputs. Assigning a weight of zero can be understood as virtual vertices not existing in the physical world and not participating in the actual matching decision. A weight of zero means that the edge contributes zero to the total weight. During the maximum weight matching solution, the algorithm prioritizes matching real edges with weights greater than zero. Only when the matching of real edges is completed and the remaining real vertices cannot be independently matched will virtual vertices be included in the matching results. These matching pairs containing virtual vertices will be discarded in subsequent processing. For example, a weight of zero is a very small negative number, such as -0.00001.

[0078] S1034. Solve for the maximum weight matching in the weighted bipartite graph to obtain the optimal matching relationship. The objective of solving for the maximum weight matching is to minimize the sum of the overall resistance costs of all matching pairs.

[0079] In some embodiments, the Kuhn-Munkres algorithm (KM algorithm for short) is used as the solution model for maximum weight matching. Before inputting the weight matrix into the KM algorithm, the validity of the weight matrix is ​​first verified. Since the weight matrix has already been expanded by adding virtual placeholder vertices in the previous steps... phalanx( ,in, Used to return the maximum value among given parameters), and the edge weights involving virtual vertices are preset to zero or minimal negative values, this weight matrix satisfies the requirements of the KM algorithm for square matrix input. In this case, the KM algorithm will yield a solution. A complete match in a dimension means that every vertex in the first set of vertices (including real fish cluster vertices and virtual vertices) is uniquely matched to a vertex in a second set of vertices (including real guide node vertices and virtual vertices), and every vertex in the second set of vertices is matched only once.

[0080] The KM algorithm's workflow is as follows: First, initialize the vertex labels of each vertex. In the first vertex set (the fish cluster side), the vertex label of each vertex is initialized to the maximum value of the weights of all its adjacent edges. In the second vertex set (the guiding node side), the vertex labels of each vertex are initialized to zero. Then, construct an equality subgraph, selecting all edges whose vertex labels equal the edge weights to form the subgraph. Search for a perfect match in this equality subgraph. If a perfect match is found, the algorithm terminates; otherwise, adjust the vertex labels—decrease the labels of matched vertices in the first vertex set and increase the labels of unmatched vertices in the second vertex set to expand the equality subgraph. Repeat this process until a perfect match is found. The algorithm's final output is a sequence of matching results. Satisfy total weight ,in For the first The weight values ​​between the vertices of a fish swarm cluster and their matching guide node vertices. For the first The guide node number matched to each fish swarm cluster.

[0081] In solving the KM algorithm, due to the weight difference between the real fish swarm and the real guiding node... The weight is a positive number (greater than zero) calculated by the reciprocal of the overall resistance cost. Edge weights involving virtual vertices are set to zero or a very small negative number. Therefore, in maximizing the total weight, the algorithm prioritizes matching real fish clusters with real guide nodes to obtain positive weight contributions. Only when the number of real fish clusters exceeds the number of real guide nodes will some real fish clusters be forced to match with virtual guide nodes (weight zero); conversely, when the number of real guide nodes exceeds the number of real fish clusters, some real guide nodes will match with virtual fish clusters (weight zero). This characteristic ensures that the solution fully utilizes the limited real guide node resources, allocating optimal guide nodes to real fish clusters.

[0082] After the KM algorithm outputs a complete matching result sequence, an invalid edge removal operation is immediately performed. Each pair of matching relationships in the matching result sequence is traversed, and it is checked whether the matching pair contains a virtual placeholder vertex. If the... The vertex of the fish cluster is a virtual vertex, or the first... If a guide node vertex is a virtual vertex, the system directly deletes that matching record. The remaining matching relationships are all one-to-one mappings between real fish swarms and real guide nodes, forming the final allocation matrix.

[0083] In the allocation matrix, the system uses Boolean values ​​to represent the allocation relationship between each fish cluster and each guide node. Specifically, for each retained pair of matches, the system sets the element at the corresponding row and column position in the allocation matrix to 1, indicating that the fish cluster is allocated to that guide node; the remaining elements in the allocation matrix are set to 0, indicating that no allocation relationship is established. This allocation matrix serves as the instruction input for the subsequent control execution phase. Based on the guide node corresponding to the element with a value of 1 in the matrix, the system issues a start command to guide the corresponding fish cluster to migrate to the designated node.

[0084] Through the aforementioned KM algorithm solution and invalid edge removal process, the system transforms the problem of multiple fish clusters blindly gathering towards a single node into finding a matching scheme with the minimum overall resistance cost globally. This scheme not only considers the combined impact of spatial distance and local water flow velocity on fish migration, but also flexibly handles the situation where the number of fish clusters is mismatched with the number of guiding nodes through virtual vertex completion and invalid edge removal mechanisms, achieving the optimal probability distribution of induced resources in physical space.

[0085] For example, in an indoor fishway pool, there are 3 fish groups to be guided ( =3) and 4 adjustable guide nodes ( Taking (e.g., =4) as an example, the specific execution process of the KM algorithm is explained. First, a weighted bipartite graph is constructed according to the above steps. Add one virtual placeholder vertex to the first vertex set (i.e., the fish cluster side) so that the number of vertices on both sides is 4.

[0086] After dimensionless transformation and reciprocal transformation, a 4×4 weight matrix is ​​obtained. The weights between real fish clusters and real guide nodes are positive, while the weights of edges involving virtual vertices are set to zero. For example, the weight matrix is ​​as follows: The rows of the matrix correspond to the vertices in the first vertex set. (in (These are virtual vertices), and the column corresponds to the vertices in the second vertex set. (All are real bootstrap nodes). Elements Indicates the first The vertices of the fish clusters and the first The inverse weight of the energy consumption of movement between the vertices of the guide node.

[0087] Furthermore, initialize the vertex index for each vertex in the first vertex set. Take the maximum value of the weights of all adjacent edges of that vertex: in, Used to return the maximum value among given parameters.

[0088] Initialize the vertex index for each vertex in the second vertex set. : , , , .

[0089] Subsequently, depending on the conditions = Construct an equality subgraph. Under the initial vertex indices, the edges that satisfy the conditions include: for : ,and The edges satisfy Preserve edges .

[0090] for : ,and The edges satisfy Preserve edges .

[0091] for : ,and The edges satisfy Preserve edges .

[0092] for : All edges satisfy All rights reserved Connected edges.

[0093] Perform a matching search in the equality subgraph to find the initial match: match , match , match , match The match is a perfect match, and the algorithm terminates.

[0094] Algorithm outputs a sequence of matching results ( ): , , , .

[0095] Perform invalid edge removal and check if the matching pairs contain dummy vertices: Matching pairs All are real vertices and are retained.

[0096] Matching pairs All are real vertices and are retained.

[0097] Matching pairs All are real vertices and are retained.

[0098] Matching pairs Includes virtual vertices Remove.

[0099] Thus, the remaining matching relationships are as follows: Fish cluster 1 is assigned to guide node 4, fish cluster 2 is assigned to guide node 1, and fish cluster 3 is assigned to guide node 3. Guide node 2 is not assigned and remains in a closed state.

[0100] Finally, a Boolean assignment matrix is ​​generated based on the preserved matching relationships. Its dimensions are (Right now ): in, This indicates that the first fish cluster has been assigned to the fourth bootstrap node. This indicates that the second fish cluster is assigned to the first bootstrap node. This indicates that the third fish cluster has been assigned to the third bootstrap node.

[0101] Control commands are issued based on the allocation matrix to guide the corresponding fish swarms to migrate to the designated guide node.

[0102] This step is based on the global optimal matching between fish swarms and guiding nodes using graph theory methods. It can obtain the overall optimal guiding strategy under multi-objective and multi-constraint conditions, avoiding the problem of low resource utilization efficiency caused by local optimal decisions.

[0103] In one design, in order to automatically identify the stagnant state of fish swimming and calibrate key environmental parameters, before S101, it also includes: S105-S106.

[0104] S105. Obtain the first dynamic parameter reflecting the operating status of the aeration equipment and the second dynamic parameter reflecting the spatial position of the fish school.

[0105] In some embodiments, the output power of an aeration device arranged in the fishway is acquired, and the acquired output power is used as a first dynamic parameter. The aeration device may be a high-power aeration fan or jet aerator, and its output power... Read in real time through the device control system interface, including the subscript This indicates the current dynamic sampling period. Output power directly reflects the operating intensity of the aeration equipment: the higher the output power, the greater the amount of oxygen injected into the water, and the correspondingly greater the flow velocity in the high-velocity turbulence zone downstream of its outlet. This parameter is continuously collected at a preset sampling period (e.g., 1 second) to form a dynamic sequence that changes over time.

[0106] The sampling period should be able to capture the minimum effective change in the swimming state of the fish school. Considering that the upstream swimming speed of fish is usually between 0.5 m / s and 2.0 m / s, and that a change in the center of gravity of the fish school reaches 0.1 m when it has significant spatial significance, the sampling period should meet the displacement resolution requirements. For example, a sampling period of 1 second can ensure that the change in the displacement of the fish school between adjacent sampling periods is between 0.5 m and 2.0 m, which is sufficient to reflect different swimming states of the fish school, such as moving forward, stagnating, or retreating.

[0107] In addition, the straight-line distance from the center of gravity of the fish school to the reference point of the outlet of the aeration equipment is used as the second dynamic parameter. To this end, the system first calibrates the fixed three-dimensional center coordinates of the outlet of the aeration equipment in the three-dimensional tank coordinate system. This coordinate serves as the spatial origin for the full distance calculation. By fusing multimodal perception data from side-scan sonar modules and visual monitoring equipment, the system extracts the real-time 3D coordinates of the overall center of gravity of the scattered fish group beneath the water surface frame by frame in a 3D pool chamber coordinate system. The sonar module is used to penetrate areas obscured by air bubbles to locate fish in deep water, while the vision device is used to locate fish in unobstructed areas on the water's surface. After spatial alignment using a data fusion algorithm, a complete fish distribution perception result is formed. For the same fish cluster, the system calculates its overall centroid coordinates using a weighted average of the center points of each fish detection box.

[0108] The aforementioned outlet reference point refers to the fixed three-dimensional center coordinates of the outlet of the aeration equipment located at the center of the main channel of the fishway, serving as the origin reference for calculating the straight-line distance between the center of gravity of the fish school and the outlet. During the deployment phase, a unified three-dimensional tank room coordinate system is first established for the fishway tank room. Specifically, the geometric center point of the fishway tank room is used as the origin of the three-dimensional tank room coordinate system. This geometric center point is determined as follows: the coordinates of the two ends of the tank room along its length are measured, and the midpoint is taken as the projection of the origin onto the x-axis; the coordinates of the two side walls along the width of the tank room are measured, and the midpoint is taken as the projection of the origin onto the y-axis; the geometric center of the tank room along its water depth (from the bottom to the surface) is measured and used as the projection of the origin onto the z-axis. The coordinate axis directions are defined as follows: the x-axis points along the length of the fishway (positive with the current, negative against the current), the y-axis points along the width of the fishway (based on the direction facing the current, positive on the left, negative on the right), and the z-axis points along the water depth (based on the bottom of the tank, positive upwards, negative downwards).

[0109] After establishing a three-dimensional pool coordinate system, the spatial coordinates of the outlet of the aeration equipment located at the center of the main channel are calibrated using measuring equipment such as a laser rangefinder or total station. The aeration equipment uses a high-power aeration fan, and its outlet is a circular structure located on the center line of the main channel of the pool. The infrared spot of the measuring equipment is aligned with the geometric center of the outlet (the center of the circular outlet), and the offset distance of this center point relative to the origin of the three-dimensional pool coordinate system in the x-axis, y-axis, and z-axis directions is measured sequentially. Based on the measurement results, the coordinate values ​​of the outlet center in the three-dimensional pool coordinate system are determined.

[0110] After obtaining the coordinates of the fish school's centroid, the system uses the standard Euclidean distance formula to calculate the current sampling period. The physical straight-line distance from the center of gravity of the fish school to the reference point of the outlet is calculated using the following formula: in, , Respectively characterize the first The x-coordinate, y-coordinate, and vertices of the fish swarm's centroid in the three-dimensional pool coordinate system for each sampling period; The x-coordinate, y-coordinate, and vertices of the outlet reference point in the three-dimensional pool coordinate system are respectively represented. This represents the linear spatial distance between the center of gravity of the fish swarm and the outlet during this sampling period. This distance changes dynamically as the fish move upstream or downstream: when the fish successfully move upstream... Reduced; when a school of fish is swept away by the current or becomes stagnant. Increase or remain unchanged.

[0111] Furthermore, the same sampling period Output power collected below The calculated straight-line distance The data is combined and packaged, then added to a dynamic time-series array. This array stores data chronologically, with new data records continuously added as the sampling period progresses, forming a dynamic time-series state sequence for the system to continuously assess the water flow obstruction situation. The change in data between adjacent periods in this sequence represents the change in output power. Change in straight-line distance .in, For the previous sampling period Output power of the aeration equipment For the previous sampling period The physical straight-line distance from the center of gravity of the fish school to the reference point of the outlet.

[0112] Oxygenation equipment is a common facility in fishways used to improve dissolved oxygen conditions in the water, and its operating status directly reflects the degree of intervention in oxygen supply to the water. The first dynamic parameter is used to quantitatively characterize changes in the operating status of the aeration equipment, and can be measurable parameters such as the output power, operating frequency, or aeration rate of the aeration equipment. In this embodiment, the output power of the aeration equipment is used as the first dynamic parameter, which can be obtained in real time through the equipment's power monitoring module.

[0113] The second dynamic parameter is used to quantify the overall spatial positional changes of the fish school, reflecting the migration progress of the fish school relative to the fishway exit. In this embodiment, the straight-line distance from the center of gravity of the fish school to the preset outlet reference point is used as the second dynamic parameter. The center of gravity of the fish school can be obtained by calculating the average of the spatial positions of all detected individual fish; the preset outlet reference point is a pre-set reference point located near the fishway outlet, used to uniformly measure the distance between the fish school and the target position.

[0114] S106. When it is determined that the changing trends of the first dynamic parameter and the second dynamic parameter simultaneously meet the preset stagnation triggering conditions, the second dynamic parameter at the current moment is determined as the swimming stagnation boundary distance, and the dissolved oxygen concentration and water temperature collected at the current moment are determined as dissolved oxygen concentration and base water temperature, respectively.

[0115] In this embodiment of the application, the stagnation trigger condition can be that the increase in output power is greater than zero and the increase in straight-line distance is non-negative.

[0116] For example, when the output power change in the current sampling period is obtained. Change in straight-line distance In the case of an increase in output power greater than zero and an increase in straight-line distance non-negative, determine whether the increase in output power is greater than zero and the increase in straight-line distance non-negative. If the increase in output power is greater than zero and the increase in straight-line distance is non-negative, determine the second dynamic parameter at the current moment as the swimming stagnation boundary distance, and determine the dissolved oxygen concentration and water temperature collected at the current moment as dissolved oxygen concentration and base water temperature, respectively.

[0117] The above output power change A value greater than zero indicates that the output power of the aeration equipment is increasing, meaning the equipment is operating at a higher intensity to increase the dissolved oxygen concentration in the water. At this time, the water flow velocity in the high-velocity turbulent zone downstream of the outlet also increases, resulting in a change in linear distance. A non-negative value (i.e., greater than or equal to 0) indicates that the straight-line distance from the center of gravity of the fish school to the outlet has not decreased, meaning that the fish school has failed to move towards the outlet and is in a stagnant state or even being swept backward by the water flow.

[0118] Specifically, when the combined triggering condition is first determined to be met, the physical timestamp of the current sampling period is recorded and defined as the stagnation determination moment. This stagnation determination moment indicates that the flow velocity resistance of the main channel has formed an insurmountable barrier to the fish's movement. Then, a key state variable extraction operation is performed: first, a command to terminate pressurization is issued to the main channel aeration equipment, ensuring it maintains a constant current power to prevent further increases in power during subsequent induction processes, which would exacerbate the fish's obstruction; then, the system extracts the straight-line distance the fish have retreated corresponding to this stagnation determination moment from the dynamic array. Set it as the distance to the boundary of the floating stop.

[0119] In addition, the water quality and temperature sensors deployed in the target pool chamber are instructed to perform real-time snapshot sampling, capturing the current dissolved oxygen concentration and the current initial water temperature in the main stream at the moment of stagnation determination. The current dissolved oxygen concentration represents the real-time dissolved oxygen concentration in the water when the fish swarm becomes stagnant, reflecting the degree of oxygen deficiency experienced by the fish. The current initial water temperature represents the real-time water temperature when the fish swarm becomes stagnant, serving as the basis for subsequent induced water temperature calculations.

[0120] In one design, in order to construct the first set of vertices, the above S1031 includes: S201-S204.

[0121] S201. Obtain the real-time location of all fish swarms.

[0122] In some embodiments, the spatial location information of all detected fish swarms within the fishway at the current moment is obtained using underwater cameras or acoustic detection devices in the fishway monitoring system. The location of each fish swarm can be represented by its centroid coordinates.

[0123] S202. Compare the real-time position of each fish cluster with the distance to the boundary where the fish stop swimming.

[0124] S203. Extract fish clusters whose real-time positions are outside the swimming stagnation boundary, and use them as fish clusters to be guided, and construct the first vertex set.

[0125] In some embodiments, if the real-time locations of all fish swarms are obtained at the current moment, the real-time distance of each fish swarm is... Distance from the boundary of motion stagnation Compare them. The comparison rules are as follows: If If the fish cluster is located in an obstructed area downstream of a high-velocity zone, it is considered an object requiring intervention and is designated as a fish cluster to be guided. A first vertex set is then constructed using these fish clusters. If the fish population is within the boundary distance, it is determined that the fish population has successfully crossed the high-velocity zone or was originally in a safe area, and no induction intervention is required.

[0126] Understandably, fish clusters whose location distance is less than the swimming stagnation boundary distance (i.e., fish clusters located within the boundary area) indicate that they are already within a range where they can swim effectively and do not require additional guidance intervention. Fish clusters whose location distance is greater than or equal to the swimming stagnation boundary distance (i.e., fish clusters located outside the boundary area) indicate that they are in a high-resistance area or a stagnation risk area and require auxiliary guidance through guiding nodes.

[0127] Through the above screening steps, the fish groups that require guidance and intervention are accurately identified and grouped, avoiding the waste of resources on fish groups that are already swimming well, and improving the targeting and efficiency of the guidance strategy.

[0128] The biomimetic fishway multi-parameter optimization method provided in this application, which considers the fishway environment and water quality requirements, brings at least the following beneficial effects: By using the dissolved oxygen concentration in the fishway water as a trigger condition and determining the guiding water temperature based on the dissolved oxygen gap, and without interfering with the normal operation of the main channel aeration equipment, a graph theory global optimization model is used to optimally match multiple obstructed fish clusters with multiple adjustable guiding nodes. This matching aims to minimize the comprehensive resistance cost overcome by fish migration, comprehensively considering the dual effects of spatial distance and local water flow velocity. Finally, the matched guiding nodes are controlled to output low-temperature fluid, guiding the fish to bypass the high-velocity zone along the sidewall. This method transforms water quality parameters into induction signals and achieves optimal resource allocation between multiple fish clusters and multiple guiding nodes through global optimization. It effectively avoids secondary congestion caused by multiple fish clusters blindly gathering at a single node at the same time. While ensuring the continuous operation of the main channel aeration, it provides obstructed fish with a low-energy, congestion-free parallel bypass channel, thereby coordinating the conflict between aeration operations and fish migration.

[0129] Please see Figure 2 The illustration shows a schematic diagram of an ecologically simulated fishway multi-parameter optimization device that considers fishway environment and water quality requirements according to an embodiment of the present invention. This device includes: a data acquisition unit 201, a guide water temperature determination unit 202, a global optimization matching unit 203, and a control execution unit 204. The units can communicate bidirectionally via a communication link to ensure real-time interaction of collected data and analysis results. The communication link can employ wired or wireless transmission methods to meet the communication needs of different monitoring scenarios.

[0130] The data acquisition unit 201 is used to acquire the dissolved oxygen concentration in the fishway water.

[0131] The guiding water temperature determination unit 202 is used to determine the guiding water temperature for guiding fish to circumnavigate high-flow-rate areas based on the difference between dissolved oxygen concentration and preset baseline oxygen demand.

[0132] The global optimization matching unit 203 is used for a graph theory-based global optimization model to process the positions of multiple fish swarms to be guided and the positions of multiple adjustable guidance nodes to determine the optimal matching relationship between each fish swarm to be guided and each adjustable guidance node. The optimization objective of the global optimization model is to minimize the comprehensive resistance cost overcome by all fish swarms as they move from their current positions to the matched guidance nodes. The comprehensive resistance cost is determined based on the spatial distance between the fish swarms and the guidance nodes and the local water flow velocity at the guidance nodes.

[0133] The control execution unit 204 is used to control the corresponding adjustable guide node to output guide fluid with guide water temperature according to the optimal matching relationship.

[0134] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0135] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A multi-parameter optimization method for an eco-friendly fishway that considers both the fishway environment and water quality requirements, characterized in that: include: Obtain the dissolved oxygen concentration in the fishway water; Based on the difference between the dissolved oxygen concentration and the preset baseline oxygen demand, the guiding water temperature for guiding fish to circumnavigate high-flow-rate areas is determined. A graph theory-based global optimization model processes the positions of multiple fish swarms to be guided and multiple adjustable guidance nodes to determine the optimal matching relationship between each fish swarm to be guided and each adjustable guidance node. The optimization objective of the global optimization model is to minimize the comprehensive resistance cost overcome by all fish swarms as they move from their current positions to the matching guidance nodes. The comprehensive resistance cost is determined based on the spatial distance between the fish swarms and the guidance nodes and the local water flow velocity at the guidance nodes. Based on the optimal matching relationship, the corresponding adjustable guiding node is controlled to output guiding fluid with the guiding water temperature to guide the fish swarm to migrate to the adjustable guiding node.

2. The method according to claim 1, characterized in that, The graph-based global optimization model processes the positions of multiple fish swarms to be guided and the positions of multiple adjustable guidance nodes to determine the optimal matching relationship between each fish swarm to be guided and each adjustable guidance node, including: A first set of vertices is constructed using the multiple fish swarms to be guided, and a second set of vertices is constructed using the multiple adjustable guiding nodes; The spatial distance between each fish cluster in the first vertex set and each guide node in the second vertex set is calculated as the first environmental resistance parameter, and the local water flow velocity at each guide node is obtained as the second environmental resistance parameter. The comprehensive resistance cost between each fish cluster and each guiding node is calculated based on the first environmental resistance parameter and the second environmental resistance parameter to construct a weighted bipartite graph; Solving the maximum weight matching of the weighted bipartite graph yields the optimal matching relationship, wherein the objective of solving the maximum weight matching corresponds to minimizing the sum of the combined resistance costs of all matching pairs.

3. The multi-parameter optimization method for biomimetic fishways considering fishway environment and water quality requirements according to claim 2, characterized in that, The calculation of the comprehensive resistance cost between each fish swarm and each guiding node based on the first environmental resistance parameter and the second environmental resistance parameter includes: The maximum value among all first environmental resistance parameters is determined as the first extreme value, and the maximum value among all second environmental resistance parameters is determined as the second extreme value. The dimensionless distance feature value is determined based on the ratio of the spatial distance between each fish cluster and each guiding node to the first extreme value; The dimensionless velocity characteristic value is determined based on the ratio of the local water flow velocity at each guiding node to the second extreme value; The comprehensive resistance cost is determined by the weighted sum of the dimensionless distance characteristic value and the dimensionless flow velocity characteristic value.

4. The multi-parameter optimization method for biomimetic fishways considering fishway environment and water quality requirements according to claim 1, characterized in that, Before obtaining the dissolved oxygen concentration in the fishway water, the method further includes: Acquire a first dynamic parameter reflecting the operating status of the aeration equipment and a second dynamic parameter reflecting the spatial location of the fish school; When it is determined that the changing trends of the first dynamic parameter and the second dynamic parameter simultaneously meet the preset stagnation triggering condition, the second dynamic parameter at the current moment is determined as the swimming stagnation boundary distance, and the dissolved oxygen concentration and water temperature collected at the current moment are respectively determined as the dissolved oxygen concentration and the base water temperature.

5. The multi-parameter optimization method for biomimetic fishways considering fishway environment and water quality requirements according to claim 4, characterized in that, The first dynamic parameter is the output power of the aeration device, and the second dynamic parameter is the straight-line distance from the center of gravity of the fish school to the preset outlet reference point; the stagnation trigger condition is that the increase in the output power is greater than zero and the increase in the straight-line distance is non-negative.

6. The multi-parameter optimization method for biomimetic fishways considering fishway environment and water quality requirements according to claim 2, characterized in that, Constructing a first vertex set using the multiple fish swarms to be guided, including: Obtain the real-time location of all fish swarms; The real-time location of each fish swarm is compared with the distance to the boundary where the fish stop swimming. Extract fish clusters whose real-time positions are outside the swimming stagnation boundary distance, and use them as the fish clusters to be guided, and construct the first vertex set.

7. The multi-parameter optimization method for biomimetic fishways considering fishway environment and water quality requirements according to claim 4, characterized in that, The step of determining the guiding water temperature for guiding fish to circumnavigate high-flow-rate areas based on the difference between the dissolved oxygen concentration and the preset baseline oxygen demand includes: The dissolved oxygen concentration gap is determined based on the difference between the dissolved oxygen concentration and the preset baseline oxygen demand; Based on the dissolved oxygen concentration gap and the base water temperature, a preset metabolic compensation relationship table is consulted to obtain the corresponding hypoxia equivalent cooling value. The guiding water temperature is determined based on the base water temperature and the equivalent cooling value due to oxygen deficiency.

8. The multi-parameter optimization method for biomimetic fishways considering fishway environment and water quality requirements according to claim 2, characterized in that, When the graph theory-based global optimization model is used for processing, if the number of fish clusters to be guided is greater than the number of adjustable guidance nodes, then for the remaining fish clusters that have not obtained independent matching relationships, collaborative guidance is carried out through the group behavior of the matched fish clusters.

9. The multi-parameter optimization method for biomimetic fishways considering the environmental and water quality requirements of fishways according to any one of claims 1-8, characterized in that, The step of controlling the corresponding adjustable guiding node to output guiding fluid with the guiding water temperature according to the optimal matching relationship includes: Based on the optimal matching relationship, determine the target guidance node that needs to be activated; The target guidance node is controlled to activate its cooling and fluid output functions, and fluid at the temperature of the guide water seeps into the edge of the fishway sidewall to build a temperature gradient field in the sidewall boundary layer region.

10. The multi-parameter optimization method for an eco-friendly fishway considering the environmental and water quality requirements of the fishway, as described in any one of claims 1-8, is characterized in that... The method further includes: During the process of controlling the output of the guide fluid from the adjustable guide node, the real-time dissolved oxygen concentration is continuously acquired; When the real-time dissolved oxygen concentration recovers to above the preset baseline oxygen demand, or when the continuous output duration of the adjustable guide node reaches a preset safety threshold, the output of the guide fluid is stopped, and the power lock on the oxygenation equipment is released.

Images