Prediction method for evaluating fish passing effect of fish lifting machine

Abstract

The invention discloses a prediction method for evaluating the fish passing effect of a fish lifting machine. The prediction method comprises the following steps: step 1, monitoring the cluster effect of fishes under a dam; 2, a releasing experiment is marked in an inlet area of the fish lifting machine; 3, constructing a fish passing effect evaluation index system; 4, establishing a fish passing effect evaluation model; 5, key factors influencing the fish passing effect of the fish lifting machine are recognized; 6, predicting and evaluating the fish passing efficiency of the fish lifting machine based on the optimal model; according to the method, the fish dam-passing behavior data is obtained through combination of ultrasonic tracking and PIT marking, a multi-factor coupled fish-passing effect prediction model is constructed, the attraction efficiency and passing efficiency of the fish lifting machine to fishes can be quantitatively evaluated, and a scientific quantitative basis is provided for operation management and optimization of the fish lifting machine.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of ecological water conservancy and fish protection, and particularly relates to a prediction method for evaluating the fish lifting effect of a fish elevator. BACKGROUND

[0002] The connectivity of river ecosystems is an important basis for maintaining aquatic biodiversity and ecological balance. In recent years, with the large-scale construction of water conservancy projects worldwide, a large number of dams and sluices have blocked the migration channels of fish, severely damaging the original water ecological connectivity, leading to a sharp decline in fish populations, and even the risk of endangerment or extinction. In order to effectively alleviate the ecological impact caused by water conservancy facilities, countries have successively carried out research and construction of fish passage facilities, helping fish to cross engineering obstacles through artificial channels. The fish elevator, as a special mechanical fish passage facility, is particularly suitable for high-drop and large-flow water conservancy engineering scenarios, and has become an important research and application hotspot in recent years. At present, developed countries such as Europe and the United States have carried out systematic and in-depth research on fish elevators, and have made extensive explorations on the optimization of design parameters, operation management and effect evaluation of fish elevators through advanced fish behavior monitoring and tracking technology, numerical simulation and statistical modeling methods, and have achieved mature research results.

[0003] In contrast, China started late in the research of fish elevator fish passage effect, and is currently in the stage of gradual exploration. Certain fish elevator engineering application practices have been carried out, and a monitoring system based on acoustic telemetry, video monitoring and radio frequency identification technology has been initially established, and a batch of preliminary fish elevator operation data has been obtained. However, the current domestic research mainly focuses on the basic observation of fish elevator operation conditions and single-factor influence analysis, and lacks a systematic evaluation method for the coupling effect of multiple factors such as fish biological characteristics, environmental parameters, fish elevator structure and operation conditions. At the same time, the existing evaluation methods still rely on simple statistics and direct observation methods, which are difficult to fully depict the nonlinear interaction effects of fish behavior and environmental factors under the complex operation mechanism of the fish elevator, and are also difficult to quantitatively analyze the contribution degree of each factor to the fish passage efficiency of the fish elevator. The above shortcomings directly limit the further improvement of the fish elevator technical system in China, and also restrict the fine operation management and facility optimization of the fish elevator engineering.

[0004] Therefore, how to comprehensively consider fish biological factors and multiple environmental factors, establish a scientific and reasonable fish passage effect prediction model, and accurately evaluate and predict the operation effect of the fish elevator has become an important problem that needs to be solved in the ecological protection engineering in China at present. SUMMARY

[0005] The technical problem solved by the present application is to provide a prediction method applied to evaluating fish passing effect of fish lift, which can quantitatively evaluate the attraction efficiency and passing efficiency of fish lift on fish, and provide scientific quantitative basis for operation management and optimization of fish lift.

[0006] To solve the above technical problem, the technical solution adopted by the present application is: a prediction method applied to evaluating fish passing effect of fish lift, which comprises the following steps:

[0007] Step 1: monitoring fish cluster effect under dam;

[0008] Step 2: marking and releasing experiment in the entrance area of fish lift;

[0009] Step 3: constructing fish passing effect evaluation index system;

[0010] Step 4: establishing fish passing effect evaluation model;

[0011] Step 5: identifying key factors affecting fish passing effect of fish lift;

[0012] Step 6: predicting and evaluating fish passing efficiency of fish lift based on optimal model.

[0013] Further, the specific process of step 1 is as follows:

[0014] An ultrasonic fish trajectory positioning system is arranged in the water area under the fish lift to form an underwater triangular positioning array, and the array position is recorded by GPS to obtain fish migration trajectory data, and the cluster distribution of fish under the dam is analyzed to identify the fish cluster hotspot area at the entrance of the fish lift.

[0015] Further, the specific process of step 2 is as follows:

[0016] PIT induction coils are arranged in the fish cluster hotspot area and at the entrance of the fish lift, fish are PIT marked and released, and the monitoring time and occurrence frequency data of uplink fish individuals are automatically collected through the induction coils.

[0017] Further, in step 3, the fish passing effect evaluation indexes of fish lift include attraction rate, attraction time, passing rate and passing time.

[0018] Further, each evaluation index is defined as follows:

[0019] (1) Attraction rate, defined as the ratio of the number of marked fish entering the entrance of the fish lift to the total number of released marked fish:

[0020]

[0021] η1 = n1 / n2 (1)

[0022] (2) Attraction time, defined as the time interval between the release of the tagged fish and the first detection of the tagged fish by the induction coil at the entrance of the fish elevator:

[0023] ET = T0-T1 (2)

[0024] where ET is the attraction time, T0 is the time at which the tagged fish is identified by the induction coil at the entrance of the fish elevator, and T1 is the time at which the tagged fish is released at the entrance area of the fish elevator;

[0025] (3) Passage rate, defined as the ratio of the number of tagged fish that successfully pass through the fish elevator to the number of tagged fish that enter the entrance of the fish elevator:

[0026]

[0027] where η2 is the passage rate, n2 is the number of tagged fish monitored by the induction coil at the entrance of the fish elevator, and n3 is the number of tagged fish that successfully pass through the fish elevator and are monitored by the induction coil at the exit of the fish elevator;

[0028] (4) Passage time, defined as the time interval between the detection of the tagged fish by the induction coil at the entrance of the fish elevator and the monitoring of the tagged fish by the induction coil at the exit of the fish elevator:

[0029] PT = T2-T0 (4)

[0030] where PT is the passage time, T2 is the time at which the tagged fish is identified by the induction coil at the exit of the fish elevator, and T0 is the time at which the tagged fish is identified by the induction coil at the entrance of the fish elevator.

[0031] Further, the specific process of step 4 is as follows:

[0032] To identify the key factors affecting the fish passage effect of the fish elevator, a fish passage effect evaluation model of the fish elevator is constructed based on the Royston-Parmar flexible parameter survival analysis model FPM. The model uses fish biological factors and environmental factors as covariates, and uses the spline function expansion of the log cumulative risk function to fit the survival distribution of the time required for fish to pass through the fish elevator. The fish biological factors include the species, body length, body weight, and gender of the fish, and the environmental factors include water temperature, flow rate, pH value, and diurnal rhythm.

[0033] Further, the expression of the fish passage effect evaluation model of the fish elevator is as follows:

[0034]

[0035] H(t i ∣xi ) represents the i-th sample at time t i cumulative risk function of the fish elevator before; s(log t i ) represents a natural cubic spline function about log time, used to flexibly fit the baseline risk; x i1 ,x i2 ,…,x ip are the values of the p covariates of the i-th sample, including fish species, body length, body weight, gender, water temperature, flow rate, pH value and circadian rhythm; β1, β2, …, β p are the regression coefficients of each main effect covariate; γ jk are the regression coefficients of the interaction term between covariates j and k; x ij ,x ik represent the values of the i-th sample on covariates j, k, respectively.

[0036] Further, the step 5 specifically processes as follows:

[0037] The optimal model is screened out by using Akaike information criterion AIC to identify the key factors affecting the fish passing effect of the fish elevator, and to quantify the influence degree of each key factor on the fish passing effect of the fish elevator, and the smaller the AIC value is, the more optimal the model is in balancing the complexity and fitting effect.

[0038] Further, the calculation of the AIC value is according to the Akaike information criterion formula:

[0039] AIC = 2K - 2lnL (θ) (6)

[0040] Wherein K is the number of model parameters; lnL (θ) is the value of the log-likelihood function of the model.

[0041] Further, the step 6 specifically processes as follows:

[0042] The key habitat factor and the cumulative function curve of the passing time of the target fish are drawn, the optimal prediction model is used to simulate and predict the fish passing effect of the fish elevator, by inputting the data of different fish population characteristics and environmental conditions, the evaluation indexes such as the attraction rate, the passing rate and the passing time of the fish elevator under the corresponding conditions are output, and the operation effect of the fish elevator is evaluated according to the evaluation indexes, so as to provide a quantitative basis for formulating the operation and management measures and optimization scheme of the fish elevator.

[0043] The present application has the following beneficial effects:

[0044] 1. The application establishes a systematic and quantitative fish elevator fish passing effect evaluation method, combines ultrasonic fish trajectory positioning system and PIT marking release technology, and first realizes accurate identification of dam downstream cluster migration behavior and continuous dynamic monitoring of fish individual fish passing process. The method breaks through the technical bottlenecks of traditional fishway evaluation methods in low spatial and temporal resolution, poor individual tracking ability and insufficient data, and significantly improves the scientificity and practicability of fish elevator fish passing efficiency evaluation.

[0045] 2. The application first introduces the Royston-Parmar flexible parameter survival model into the ecological evaluation of the fish elevator, constructs a multi-factor evaluation model combining fish biological characteristics and water environmental factors, and can effectively describe the nonlinear relationship between fish passing time distribution and key influencing factors. Through AIC criterion to optimize the model structure, not only improve the prediction accuracy, but also clarify the dominant role of each factor in different fish passing behaviors, provide data support and decision basis for subsequent fish elevator operation strategy optimization.

[0046] 3. Through the implementation of the application, the fish passing response characteristics of different fish populations under specific hydrological and ecological conditions can be accurately identified, the adaptability design and ecological function evaluation ability of the fish elevator are comprehensively improved, which is helpful to build a scientific and sustainable fish migration channel system, and provides key technical support for the planning and design of ecological fish passing facilities in the Yangtze River and other dam water systems and ecological restoration.

[0047] 4. The application obtains fish passing behavior data by combining ultrasonic tracking with PIT marking, constructs a multi-factor coupled fish passing effect prediction model, and can quantitatively evaluate the attraction efficiency and passing efficiency of the fish elevator to the fish, and provide scientific and quantitative basis for the operation management and optimization of the fish elevator. DETAILED DESCRIPTION

[0048] Figure 1 It is a step flow chart of the prediction method for evaluating the fish passing effect of the fish elevator;

[0049] Figure 2 It is a dam downstream fish cluster distribution hotspot map;

[0050] Figure 3 It is a dependent variable matrix correlation analysis graph of the fish passing effect evaluation model of the fish elevator;

[0051] Figure 4 It is a D+Q+T optimal model parameter evaluation forest graph;

[0052] Figure 5 It is a time-circadian rhythm cumulative function graph based on the Royston-Parmar passing rate model;

[0053] Figure 6This is a graph of the time-flow accumulation function based on the Royston-Parmar throughput model; Detailed Implementation

[0054] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0055] Example 1: As Figures 1 to 6 As shown, this embodiment provides a predictive method for evaluating the fish passage effect of a fish lift, aiming to comprehensively improve the monitoring and protection level of fish resources in river damming areas. By combining underwater acoustic monitoring, telemetry tagging, and other technologies, a comprehensive monitoring system covering the activity waters of target fish is constructed, achieving multi-dimensional real-time monitoring of fish migration trajectories and habitat characteristics. Based on monitoring data, the main activity hotspots of fish downstream of the dam are accurately identified, revealing the spatiotemporal distribution patterns of fish resources. Key indicators of suitable fish habitats (such as water temperature, flow velocity, pH value, and diurnal rhythms) are further collected, and a dynamic evaluation model of fish passage effect coupling fish biological factors and environmental factors is constructed. This model identifies key factors affecting the fish passage effect and predicts the trend of fish passage efficiency changes under the combined effect of multiple environmental factors, ultimately proposing fish lift operation management measures and optimization schemes based on scientific prediction. This embodiment uses multi-source monitoring technology as its core, systematically integrating multiple monitoring methods, which not only significantly improves monitoring accuracy and efficiency but also provides comprehensive scientific support for the evaluation of fish passage effects and resource protection. The steps are as follows:

[0056] Step 1: Monitoring the fish herding effect downstream of the dam;

[0057] Step 2: Marking and releasing experiment at the inlet area of ​​the fish lift;

[0058] Step 3: Construct an evaluation index system for fish passage effectiveness;

[0059] Step 4: Establish a fish passage effect evaluation model;

[0060] Step 5: Identify the key factors affecting the fish lifter's fish-passing performance;

[0061] Step 6: Evaluate the efficiency of the fish lifter based on the optimal model.

[0062] In step 1, an ultrasonic fish trajectory positioning system is deployed in the waters below the fish lift dam to form an underwater triangular positioning array. The array location is recorded using GPS to obtain fish migration trajectory data and analyze the fish cluster distribution below the dam to identify fish cluster hotspots at the entrance of the fish lift.

[0063] In step 2, PIT induction coils are arranged in the fish cluster hotspot area and at the entrance of the fish elevator, PIT marking is performed on the fish and the fish is released, and the monitoring time and occurrence frequency of the upgoing fish individuals are automatically collected through the induction coils.

[0064] In step 3, the fish elevator fish passing effect evaluation indexes include attraction rate, attraction time, passing rate and passing time, and each evaluation index is defined as follows:

[0065] (1) Attraction rate, defined as the ratio of the number of marked fish entering the entrance of the fish elevator to the total number of released marked fish:

[0066]

[0067] Where η1 is the attraction rate; n1 is the number of marked fish released in the entrance area of the fish elevator; n2 is the number of marked fish monitored by the induction coil at the entrance of the fish elevator.

[0068] (2) Attraction time, defined as the time interval between the release of the marked fish and the first detection of the marked fish by the entrance of the fish elevator:

[0069] ET = T0-T1 (2)

[0070] Where ET is the attraction time; T0 is the time when the induction coil at the entrance of the fish elevator recognizes the marked fish; T1 is the time when the marked fish is released in the area of the entrance of the fish elevator.

[0071] (3) Passing rate, defined as the ratio of the number of marked fish successfully passing through the fish elevator to the number of marked fish entering the entrance of the fish elevator:

[0072]

[0073] Where η2 is the passing rate, n2 is the number of marked fish monitored by the induction coil at the entrance of the fish elevator; n3 is the number of marked fish successfully passing through the fish elevator monitored by the induction coil at the outlet of the fish elevator.

[0074] (4) Passing time, defined as the time interval between the detection of the marked fish by the entrance of the fish elevator and the monitoring of the marked fish by the outlet of the fish elevator:

[0075] PT = T2-T0 (4)

[0076] Where PT is the passing time, T2 is the time when the induction coil at the outlet of the fish elevator recognizes the marked fish; T0 is the time when the induction coil at the entrance of the fish elevator recognizes the marked fish.

[0077] In step 4, to identify the key factors affecting the fish lift's fish passage efficiency, a fish lift efficiency evaluation model based on the Royston-Parmar Flexible Parametric Survival Analysis (FPM) model was constructed. This model uses fish biological factors (including fish weight, body length, body width, body height, and condition factor) and environmental factors (including reservoir water level, turbidity, recovery time after tag implantation, water temperature, flow rate, diurnal rhythm, and dissolved oxygen) as covariates. It uses the spline function expansion of the log-cumulative hazard function to fit the survival distribution of the time required for fish to pass through the fish lift. Examples of data file variables in the model are shown in Table 1.

[0078] Table 1. Data file variables for the Royston-Parmar flexible parametric survival analysis model.

[0079]

[0080] The expression for the fish lift's fish passage effect evaluation model is as follows:

[0081]

[0082] In the formula: H(t) i |x i ) represents the i-th sample at time t i The cumulative risk function of the fish-lifting machine; s(log t) i ) represents a natural cubic spline function with respect to logarithmic time, used for flexible fitting of baseline risk; x i1 ,x i2 ,…,x ip Let p be the values ​​of the p covariates for the i-th sample, including fish species, body length, weight, sex, water temperature, current velocity, pH value, and diurnal rhythm; β1, β2, ..., β p γ represents the regression coefficients of each main effect covariate; jk x represents the regression coefficient of the interaction term between covariates j and k; ij ,x ik Let represent the values ​​of the i-th sample in covariates j and k, respectively.

[0083] In step 5, to address the potential multicollinearity problem of covariates in the fish lifter performance evaluation model constructed in step 4 and to improve the model's prediction accuracy, the Akaike Information Criterion (AIC) is used to select the optimal model, identify the key factors affecting the fish lifter performance, and quantify the degree of influence of each key factor on the fish lifter performance. The smaller the AIC value, the better the model is in balancing complexity and fitting effect.

[0084] The AIC value is calculated according to the Akaike Information Criterion Formula:

[0085] AIC = 2K - 2lnL(0) (6)

[0086] wherein K is the number of model parameters; lnL(0) is the log-likelihood function value of the model.

[0087] In step 6, the key habitat factors and the target fish passage time cumulative function curve are drawn, the optimal prediction model is used to simulate and predict the fish ladder passage effect, by inputting the data of different fish population characteristics and environmental conditions, the evaluation indexes such as the attraction rate, passage rate and passage time of the fish ladder under the corresponding conditions are output, and the operation effect of the fish ladder is evaluated, thereby providing a quantitative basis for formulating the operation and management measures and optimization scheme of the fish ladder.

[0088] Example 2: SWL hydropower station is located in the upper reaches of the Jinsha River, and is the first large-scale hydropower project with a capacity of 1 million kilowatts on the Tibetan Plateau. For the fish ladder system matched with the dam body of the hydropower station, the present application takes the dam downstream area of the SWL hydropower station as an example to provide a specific prediction and evaluation calculation example, and verifies the scientificity and operability of the fish ladder passage effect evaluation method. The specific implementation process is as follows:

[0089] 1. An ultrasonic fish trajectory positioning system is arranged in the typical water area downstream of the SWL hydropower station to form a multi-point underwater positioning triangular array, and a GPS device is used to record the three-dimensional spatial coordinates of each receiving unit to realize accurate tracking of the migration path of fish. The monitoring database is established by combining the environmental parameters (including reservoir water level, turbidity, recovery time after implanting tags, water temperature, flow, diurnal rhythm and dissolved oxygen value) during the monitoring period and the biological information of fish (such as body weight, body length, body width, body height and fatness, etc.). Through time series analysis, the distribution characteristics of the cluster hot spot area of fish in the dam downstream area are obtained, and the basis for the distribution of the marked release experiment is provided (see Figure 2 ).

[0090] 2. According to the cluster hot spot area identified in step 1, a PIT induction coil system is arranged in the range of about 3-5 meters in front of the fish ladder inlet, and target fish are selected in the hot spot area for PIT marking and release experiment. The time, frequency and residence behavior of fish entering and leaving the fish ladder are automatically recorded by the induction coil to form individual fish tracking data, which are used to calculate the attraction rate, attraction time and other indexes (see Table 2).

[0091] 3. The attraction rate and attraction time are collected as the fish passage effect evaluation indexes of the fish ladder inlet attraction rate model. Before the model is constructed, the correlation analysis graph of the dependent variable matrix of the fish ladder inlet attraction rate model is drawn (see Figure 3), and the correlation between condition factor (CF) and turbidity (CD) was strong (F>0.6). Therefore, the four variables, body weight (W), body width (BW), body height (BH) and turbidity (CD), were removed from the model of the intake attraction rate of the fish elevator.

[0092] 4. Based on the Royston-Parmar flexible parameter survival model (FPM), the model of the intake attraction rate of the fish elevator was constructed. Body length (BL), reservoir level (RL), recovery time after tagging (RT), water temperature (T), flow (Q), diel rhythm (D), condition factor (CF) and dissolved oxygen (DO) were selected as independent variables, and the passing time of each PIT-tagged fish was defined as Time. The baseline hazard function was fitted using a natural cubic spline function (see Table 2).

[0093] Table 2 Data file variables of the Royston-Parmar flexible parameter survival analysis model

[0094]

[0095] 5. The Akaike information criterion (AIC) was used to screen multiple model combinations, and finally the model containing diel rhythm (D), flow (Q) and water temperature (T) as three covariates was determined to have the lowest AIC value and the best fitting effect (see Table 3). This model was used as the optimal model of the intake attraction rate of the fish elevator to identify the key factors affecting the intake attraction rate of the fish elevator.

[0096] Table 3 Model selection based on the Akaike information criterion (AIC)

[0097]

[0098] Note: w i = EXP(-0.5Δ i AIC) / [EXP(-0.5Δ1AIC)+EXP(-0.5Δ2AIC)+EXP(-0.5Δ3AIC)+EXP(-0.5Δ4AIC)+EXP(-0.5Δ5AIC)], w i is the model weight, w j is the optimal model weight.

[0099] 6. Based on the regression coefficients and significance test results of the optimal model (see Table 4), it can be seen that circadian rhythm (D) and water temperature (T) have a significant impact on the fish lift inlet attraction rate model (P < 0.05), while flow rate (Q) has a highly significant impact on the fish lift inlet attraction rate (P < 0.01). Circadian rhythm (D), flow rate (Q), and water temperature (T) are all positively correlated with the fish lift inlet attraction rate (see Table 4). Specific analysis results indicate that the labeled experimental fish prefer to pass through the fish lift inlet during the day (see Table 4). Figure 5 When the flow rate increases by 1m 3 At a flow rate of / s, the probability of the labeled test fish passing through the fish lift inlet increased by 25.1%, but under high flow conditions, the time for fish to pass through the fish lift inlet was longer, which was not conducive to smooth passage (see...). Figure 6 For every 1°C increase in temperature, the probability of the marked test fish entering the fish lift increased by 34.9% (see Table 4).

[0100] Table 4. Optimal Model Parameter Evaluation in Royston-Parmar Flexible Parametric Survival Analysis

[0101]

[0102] * indicates significance.

[0103] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A predictive method for evaluating the fish-passing effect of a fish lifter, characterized in that: It includes the following steps: Step 1: Monitoring of fish herding effect downstream of the dam; Step 2: Marking and releasing experiment at the inlet area of ​​the fish lift; Step 3: Construct an evaluation index system for fish passage effectiveness; Step 4: Establish a fish passage effect evaluation model; Step 5: Identify the key factors affecting the fish lifter's fish-passing performance; Step 6: Evaluate the efficiency of the fish lifter based on the optimal model.

2. The prediction method for evaluating the fish-passing effect of a fish lifter according to claim 1, characterized in that: The specific process of step 1 is as follows: An ultrasonic fish trajectory positioning system was deployed in the waters below the fish lift dam to form an underwater triangular positioning array. The array location was recorded using GPS to obtain fish migration trajectory data and analyze the fish cluster distribution below the dam to identify fish cluster hotspots at the entrance of the fish lift.

3. The prediction method for evaluating the fish-passing effect of a fish lifter according to claim 1, characterized in that: The specific process of step 2 is as follows: PIT induction coils are deployed in the fish cluster hotspot area and at the entrance of the fish lifter to mark the fish with PIT and release them. The monitoring time and frequency data of individual fish passing by are automatically collected through the induction coils.

4. The predictive method for evaluating the fish-passing effect of a fish lifter according to claim 1, characterized in that: In step 3, the evaluation indicators for the fish lift's fish-passing effect include attraction rate, attraction time, pass rate, and pass time.

5. The prediction method for evaluating the fish-passing effect of a fish lifter according to claim 4, characterized in that: The evaluation indicators are defined as follows: (1) Attraction rate, defined as the ratio of the number of tagged fish entering the fish lift inlet to the total number of tagged fish released: Where η1 is the attraction rate; n1 is the number of tagged fish released in the entrance area of ​​the fish lifter; and n2 is the number of tagged fish detected by the induction coil at the entrance of the fish lifter. (2) Attraction time, defined as the time interval between the start of release of the tagged fish and its first detection at the fish lift inlet: ET = T0 - T1 (2) Where ET is the attraction time; T0 is the moment when the induction coil at the entrance of the fish lifter detects the tagged fish; and T1 is the moment when the tagged fish is released from the area at the entrance of the fish lifter. (3) Pass rate, defined as the ratio of the number of marked fish that successfully pass through the fish lift to the total number of marked fish entering the fish lift entrance: Where η2 is the pass rate, n2 is the number of marked fish detected by the induction coil at the entrance of the fish lift, and n3 is the number of marked fish that successfully passed through the fish lift detected by the induction coil at the exit of the fish lift. (4) Time, defined as the time interval between when the marked fish is detected at the inlet of the fish lifter and when it is detected at the outlet of the fish lifter: PT = T2 - T0 (4) Where PT is the passage time, T2 is the moment when the induction coil at the fish lifter exit detects the marked fish, and T0 is the moment when the induction coil at the fish lifter entrance detects the marked fish.

6. The predictive method for evaluating the fish-passing effect of a fish lifter according to claim 1, characterized in that: The specific process of step 4 is as follows: To identify the key factors affecting the fish lift's fish passage efficiency, a fish lift efficiency evaluation model based on the Royston-Parmar Flexible Parametric Survival Analysis (FPM) model was constructed. This model uses fish biological factors and environmental factors as covariates and fits the survival distribution of the time required for fish to pass through the fish lift using a spline function expansion of the log-cumulative hazard function. The fish biological factors include fish species, body length, weight, and sex, while the environmental factors include water temperature, flow rate, pH value, and diurnal rhythm.

7. The predictive method for evaluating the fish-passing effect of a fish lifter according to claim 6, characterized in that: The expression for the fish lift's fish passage effect evaluation model is as follows: In the formula: H(t) i |x i ) represents the i-th sample at time t i The cumulative risk function of the fish-lifting machine; s(log t) i ) represents a natural cubic spline function with respect to logarithmic time, used for flexible fitting of baseline risk; x i1 ,x i2 ,…,x ip Let p be the values ​​of the p covariates for the i-th sample, including fish species, body length, weight, sex, water temperature, current velocity, pH value, and diurnal rhythm; β1, β2, ..., β p γ represents the regression coefficients of each main effect covariate; jk x represents the regression coefficient of the interaction term between covariates j and k; ij ,x ik Let represent the values ​​of the i-th sample in covariates j and k, respectively.

8. The predictive method for evaluating the fish-passing effect of a fish lifter according to claim 1, characterized in that: The specific process of step 5 is as follows: The optimal model was selected using the Akaike Information Criterion (AIC). Key factors affecting the fish lifter's fish passage effect were identified, and the influence of each key factor on the fish lifter's fish passage effect was quantified. The smaller the AIC value, the better the model is in balancing complexity and fitting effect.

9. A predictive method for evaluating the fish-passing effect of a fish lifter according to claim 8, characterized in that: The AIC value is calculated according to the Akaike Information Criterion Formula: AIC=2K-2lnL(θ) (6) Where K is the number of model parameters; lnL(θ) is the log-likelihood function value of the model.

10. A predictive method for evaluating the fish-passing effect of a fish lifter according to claim 1, characterized in that: The specific process of step 6 is as follows: The cumulative function curves of key habitat factors and target fish passage time are plotted. The optimal prediction model is used to simulate and predict the fish passage effect of the fish lifter. By inputting data on different fish population characteristics and environmental conditions, the fish lifter's attraction rate, passage rate, and passage time under the corresponding conditions are output as evaluation indicators. Based on this, the operation effect of the fish lifter is evaluated, providing a quantitative basis for formulating fish lifter operation management measures and optimization schemes.

Images