A high-survival-rate transportation path planning method for cross-regional aquatic fry

Abstract

The present application relates to the technical field of path planning, in particular to a high-survival-rate transportation path planning method for cross-regional aquatic fry. The present application analyzes the change trend of the vertical acceleration and the dissolved oxygen concentration of water at different times to obtain a frequency band sensitive weight optimization vector of each sliding window; obtains the oxygen consumption recovery time length according to the vibration characteristics and time distribution characteristics of the average vibration spectrum vector of different sliding windows and the average oxygen consumption rate distribution; obtains the road section steady-state oxygen consumption rate of each candidate road section according to the preset vehicle suspension transfer coefficient of different preset frequency bands, the standard road surface power spectrum density of each candidate road section and the frequency band sensitive weight optimization vector of each sliding window; obtains the comprehensive road section cost of each candidate road section by combining the preset vehicle passing time length of each candidate road section to obtain the optimal transportation path. The present application considers biological metabolism and passing efficiency, accurately obtains the comprehensive road section cost and improves the effectiveness of path planning.

Description

Technical Field

[0001] This invention relates to the field of route planning technology, specifically to a method for planning high-survival-rate transportation routes for cross-regional aquatic seedlings. Background Technology

[0002] The cross-regional transportation of aquatic seedlings is a crucial link in modern fisheries logistics. During long-distance transportation, the mechanical vibrations generated by the transport vehicle are transmitted to the water tank through the vehicle body, causing wide-frequency sloshing of the water. This leads to an abnormally high metabolic rate of the seedlings, accelerates water quality deterioration, and depletes the seedlings' own energy reserves, resulting in a decrease in survival rate during transportation or delayed mortality after being placed in the pond. Currently, the transportation of high-value aquatic seedlings generally uses closed transport vehicles equipped with pure oxygen supply systems, eliminating the risks caused by oxygen deficiency or uneven reoxygenation due to water surface sloshing.

[0003] In terms of route planning, existing general navigation technologies mainly aim to optimize the shortest distance, the shortest time, or avoid congestion. However, different batches and sizes of seedlings have significant differences in their sensitivity to vibration frequency, and the recovery of the metabolic rate of organisms after experiencing violent shaking has a significant time lag. Considering only the physical traffic efficiency of vehicles while ignoring the specific tolerance characteristics of organisms to the transportation environment, it is impossible to achieve the goal of minimizing biological loss throughout the transportation process, resulting in poor route planning effectiveness. Summary of the Invention

[0004] To address the technical problem of poor route planning effectiveness due to considering only the physical traffic efficiency of vehicles while neglecting the specific tolerance characteristics of organisms to the transportation environment, the present invention aims to provide a high survival rate transportation route planning method for cross-regional aquatic seedlings. The specific technical solution adopted is as follows: This invention proposes a method for planning high-survival-rate transportation routes for cross-regional aquatic seedlings, the method comprising: Obtain the vertical acceleration of the water body and the dissolved oxygen concentration at every moment; A sliding window is constructed to traverse all time points. Based on the vertical acceleration of the water body at different time points, the power spectral density function of each sliding window is constructed. According to the distribution characteristics of the power spectral density function of each sliding window in different preset frequency bands and the trend of dissolved oxygen concentration at different time points within the corresponding preset delay window, the average vibration spectrum vector and average oxygen consumption rate of each sliding window are obtained. Based on the average oxygen consumption rate, average vibration spectrum vector, initial frequency band sensitive weight vector, and corresponding initial covariance matrix of different sliding windows, the frequency band sensitive weight optimization vector for each sliding window is obtained; based on the vibration characteristics and time distribution characteristics of the average vibration spectrum vector of different sliding windows, as well as the distribution of average oxygen consumption rate, the oxygen consumption recovery time is obtained. Based on the preset vehicle suspension transfer coefficients for different preset frequency bands, the standard pavement power spectral density for each candidate road segment, and the frequency band sensitivity weight optimization vector for each sliding window, the steady-state oxygen consumption ratio of each candidate road segment is obtained; based on the preset vehicle travel time, steady-state oxygen consumption rate, and oxygen consumption recovery time for each candidate road segment, the comprehensive road segment cost is obtained. The optimal transportation route is obtained based on the comprehensive road segment cost distribution of different candidate road segments.

[0005] Furthermore, the method for obtaining the power spectral density function includes: The power spectral density function is obtained by performing a Fourier transform on the sequence of vertical accelerations of the water body at different times within each sliding window.

[0006] Furthermore, the method for obtaining the average vibration spectrum vector and the average oxygen consumption rate includes: The integral value of the power spectral density function corresponding to each sliding window in each preset frequency band is obtained as the average vibration energy of each sliding window in each preset frequency band; The average vibration spectrum vector is formed by the column vectors of average vibration energy in different preset frequency bands from smallest to largest. Construct a straight line fitting the dissolved oxygen concentration at different times within a preset delay window for each sliding window, and obtain the absolute value of the slope of the straight line fitting the dissolved oxygen concentration as the average oxygen consumption rate.

[0007] Furthermore, the method for obtaining the frequency band sensitive weight optimization vector includes: Obtain the dot product between the initial frequency band sensitive weight vector and the average vibration spectrum vector of the next sliding window, and obtain the sum of the standard oxygen consumption rate and the dot product result as the predicted oxygen consumption rate for the next sliding window. Based on the initial covariance matrix and the average vibration spectrum vector of the next sliding window, the Kalman gain of the next sliding window is obtained. Based on the initial frequency band sensitive weight vector, the initial covariance matrix, the Kalman gain of the next sliding window, the average vibration spectrum vector, and the difference between the average oxygen consumption rate and the predicted oxygen consumption rate, the optimized frequency band sensitive weight vector and the covariance matrix of the next sliding window are obtained. Using the next sliding window as a new initial step, the frequency band sensitive weight optimization vector and covariance matrix of the next sliding window are obtained iteratively to obtain the frequency band sensitive weight optimization vector for each sliding window.

[0008] Furthermore, the methods for obtaining the frequency band sensitive weight optimization vector and the covariance matrix optimization vector include: The product of the Kalman gain of the next sliding window and the difference between the average oxygen consumption rate and the predicted oxygen consumption rate is used as the frequency band adjustment vector; the sum of the frequency band adjustment vector and the initial frequency band sensitive weight vector is used as the frequency band sensitive weight optimization vector for the next sliding window. Obtain the product vector between the Kalman gain of the next sliding window and the transpose of the average vibration spectrum vector; obtain the difference between the identity matrix and the product vector, calculate the product of the difference result and the initial covariance matrix, and add a preset forgetting factor to obtain the covariance matrix of the next sliding window.

[0009] Furthermore, the method for obtaining the oxygen recovery time includes: If the magnitude of the average vibration spectrum vector of a sliding window is less than the preset vibration threshold, the corresponding sliding window is taken as a stable sliding window; if the length of the period consisting of consecutive stable sliding windows is greater than the preset duration threshold, the difference between the average oxygen consumption rate at the beginning of the period consisting of consecutive stable sliding windows and the standard oxygen consumption rate is obtained as the relative difference. The length of the continuous and stable sliding window period is divided by the preset oxygen consumption recovery time, and a negative correlation mapping is performed to obtain the time decay coefficient; the product of the relative difference and the time decay coefficient is obtained, and the sum of the product result and the standard oxygen consumption rate is calculated to obtain the oxygen consumption exponential recovery function; the mean of the average oxygen consumption rate of the continuous and stable sliding window is obtained as the overall oxygen consumption rate. The oxygen consumption recovery time is obtained when the oxygen consumption index recovery function and the overall oxygen consumption rate are equal, based on the nonlinear least squares method.

[0010] Furthermore, the method for obtaining the steady-state oxygen consumption ratio of the road section includes: Based on the preset vehicle suspension transmission coefficients for different preset frequency bands and the standard pavement power spectral density for each candidate road segment, the frequency band water vibration spectral vector of each candidate road segment is obtained. The dot product between the frequency band sensitive weight optimization vector of each sliding window and the frequency band water vibration spectrum vector of each candidate road segment is obtained as the additional oxygen consumption increment. The ratio of the additional oxygen consumption increment to the standard oxygen consumption rate is obtained, and the sum of the ratio and the positive integer 1 is used as the steady-state oxygen consumption ratio of the road segment.

[0011] Furthermore, the method for obtaining the water vibration spectrum vector in the aforementioned frequency band includes: The product of the preset vehicle suspension transmission coefficient for different preset frequency bands and the standard pavement power spectral density for each candidate road segment is obtained as the water vibration energy of each candidate road segment in each preset frequency band. The column vector of water vibration energy in all preset frequency bands for each candidate road segment is formed according to the frequency band from smallest to largest, and is used as the frequency band water vibration spectrum vector.

[0012] Furthermore, the preset vehicle travel time, steady-state oxygen consumption rate of the road segment, and oxygen consumption recovery time are all positively correlated with the comprehensive road segment cost.

[0013] Furthermore, the method for obtaining the optimal transportation route includes: Based on the comprehensive road segment cost of different candidate road segments, an optimal path search algorithm is performed on all candidate road segments to obtain the path with the minimum cumulative sum of the comprehensive road segment costs of all candidate road segments between the transportation origin and destination, which is taken as the optimal transportation path.

[0014] The present invention has the following beneficial effects: This invention obtains the average vibration spectrum vector and average oxygen consumption rate of each sliding window based on the distribution characteristics of the power spectral density function corresponding to each sliding window in different preset frequency bands and the changing trend of dissolved oxygen concentration at different times within the corresponding preset delay window. This helps to analyze the frequency domain distribution mean of water body vibration energy and the rate of change of dissolved oxygen concentration. Based on the average oxygen consumption rate, average vibration spectrum vector, initial frequency band sensitivity weight vector, and corresponding initial covariance matrix of different sliding windows, an optimized frequency band sensitivity weight vector for each sliding window is obtained, reflecting the seedling's sensitivity to oxygen consumption in each vibration frequency band. Based on the vibration characteristics and time... By analyzing the distribution characteristics of oxygen consumption and the distribution of average oxygen consumption rate, the oxygen recovery time is obtained, quantifying the decay time characteristics of the oxygen consumption rate falling from the high oxygen consumption rate back to the baseline value. Based on the preset vehicle suspension transfer coefficients for different preset frequency bands, the standard pavement power spectral density of each candidate road segment, and the frequency band sensitive weight optimization vector for each sliding window, the steady-state oxygen consumption ratio of each candidate road segment is obtained, reflecting the metabolic ratio of vehicles traveling on the road segment. Based on the preset vehicle travel time, steady-state oxygen consumption rate, and oxygen recovery time of each candidate road segment, the comprehensive road segment cost of each candidate road segment is obtained, more comprehensively quantifying the dual comprehensive cost including biological metabolism and traffic efficiency; thus, the optimal transportation route is obtained. This invention considers biological metabolism and traffic efficiency, accurately obtains the comprehensive road segment cost, and improves the effectiveness of route planning. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of a method for planning a high survival rate transportation route for cross-regional aquatic seedlings, provided as an embodiment of the present invention.

[0017] Figure 2 A flowchart illustrating a method for obtaining the average vibration spectrum vector and average oxygen consumption rate according to an embodiment of the present invention; Figure 3 This is a flowchart illustrating a method for obtaining a frequency band sensitive weight optimization vector, as provided in one embodiment of the present invention. Detailed Implementation

[0018] The following description, in conjunction with the accompanying drawings, details a specific scheme for a cross-regional aquatic seedling high survival rate transportation route planning method provided by the present invention.

[0019] Please see Figure 1 The diagram illustrates a flowchart of a method for planning high-survival-rate transportation routes for cross-regional aquatic seedlings according to an embodiment of the present invention. The specific method includes: Step S1: Obtain the vertical acceleration of the water body and the dissolved oxygen concentration at each moment.

[0020] In the embodiments of the present invention, considering that the metabolic response of organisms has a significant time lag relative to environmental stimuli, and that instantaneous sensor data is easily interfered with by random noise, data is continuously collected. First, for live animal transport vehicles equipped with closed-loop pure oxygen supply devices, in such transport systems, oxygen is injected into the water body through controlled cylinders and microporous aeration discs. The top of the water tank is equipped with a wave deflector or a full-water transport method is adopted, thereby minimizing the interference of air reoxygenation caused by liquid surface sloshing, ensuring that the downward trend of dissolved oxygen sensor data is mainly dominated by the biological respiration and metabolism of aquatic seedlings. The vertical acceleration of the water body and dissolved oxygen concentration are collected by sensors for subsequent analysis.

[0021] It should be noted that in the embodiments of the present invention, the interval between moments is set to 1 second. In other embodiments of the present invention, the interval between moments can be set according to specific circumstances, and will not be limited or elaborated here.

[0022] Step S2: Construct a sliding window that iterates through all time points. Based on the vertical acceleration of the water body at different time points, construct the power spectral density function for each sliding window. According to the distribution characteristics of the power spectral density function corresponding to each sliding window in different preset frequency bands and the trend of dissolved oxygen concentration at different time points within the corresponding preset delay window, obtain the average vibration spectrum vector and average oxygen consumption rate of each sliding window.

[0023] Considering that instantaneous sensor data is easily affected by random noise, feature extraction is performed based on a sliding statistical time window to smooth out the high-frequency jitter of sensor readings. The sliding window is constructed to traverse all time points.

[0024] It should be noted that, in one embodiment of the present invention, the size of the sliding window is a time range consisting of 300 seconds, and the sliding step is 60 seconds; in other embodiments of the present invention, the size of the sliding window can be set according to specific circumstances, and will not be limited or described in detail here.

[0025] To quantify the energy distribution of vibrations at the same frequency, a sliding window is constructed to traverse all time points, obtaining the power spectral density function composed of the vertical acceleration of the water body at different time points within each sliding window.

[0026] Preferably, in one embodiment of the present invention, the method for obtaining the power spectral density function includes: The power spectral density function is obtained by performing a Fourier transform on the sequence of vertical accelerations of the water body at different times within each sliding window.

[0027] It should be noted that the Fourier transform converts the time domain into the frequency domain, describing the distribution of energy of the vibration signal as a function of frequency, which helps to reveal the long-term statistical characteristics and intrinsic structure of the vibration signal.

[0028] The power spectral density function describes the distribution of energy of a vibration signal as a function of frequency, which helps to quantify the vibration level of vehicle transportation. After a living organism senses vibration stimulation, the secretion of stress hormones and the respiratory rate require a certain amount of time to conduct, resulting in a lag in the change of metabolic rate compared to the occurrence of vibration excitation. Therefore, the trend of dissolved oxygen concentration at different times within a preset delay window is analyzed to quantify the oxygen consumption rate. Based on the distribution characteristics of the power spectral density function corresponding to each sliding window in different preset frequency bands and the trend of dissolved oxygen concentration at different times within the corresponding preset delay window, the average vibration spectrum vector and average oxygen consumption rate of each sliding window are obtained.

[0029] Preferably, in one embodiment of the present invention, the method for obtaining the average vibration spectrum vector and the average oxygen consumption rate is described in [reference needed]. Figure 2 It shows a flowchart of a method for obtaining the average vibration spectrum vector and the average oxygen consumption rate, including: Step S201: Obtain the integral value of the power spectral density function corresponding to each sliding window in each preset frequency band, as the average vibration energy of each sliding window in each preset frequency band; and form a column vector of the average vibration energy of different preset frequency bands in ascending order, as the average vibration spectrum vector.

[0030] Based on this, the average vibration energy indicates the frequency vibration energy excited by the road surface when the vehicle is driving within a certain frequency band. The greater the average vibration energy, the stronger the vibration.

[0031] It should be noted that, in one embodiment of the present invention, in order to reduce the data dimensionality, the method for obtaining the preset frequency band is as follows: based on the biological vibration sensitive range of aquatic seedlings and the inherent vibration characteristics of vehicles in the prior art, an effective vibration frequency range is obtained in advance; the system divides the effective vibration frequency range into multiple non-overlapping frequency bands for analysis, and the number of frequency bands is set to 10; in other embodiments of the present invention, the frequency bands can be set according to specific circumstances, which will not be elaborated here.

[0032] Step S202: Construct a straight line fitting the dissolved oxygen concentration at different times within the preset delay window corresponding to each sliding window, and obtain the absolute value of the slope of the straight line fitting the dissolved oxygen concentration as the average oxygen consumption rate.

[0033] It should be noted that, in the embodiments of the present invention, straight line fitting can be performed by existing fitting algorithms such as least squares method or polynomial fitting. The specific means are well known to those skilled in the art and will not be described in detail here.

[0034] It should be noted that, in the embodiments of the present invention, the slope reflects the trend of dissolved oxygen concentration change. The slope of the dissolved oxygen concentration fitting line can be obtained by taking the derivative, or by calculating the difference between the dissolved oxygen concentration at the end time and the beginning time and the interval between the time points, where the interval between the time points is the absolute value of the difference between the time points. The specific means are well known to those skilled in the art and will not be described in detail here.

[0035] It should be noted that when an organism senses vibration stimulation, the body's emergency secretion and increased respiratory rate require a certain amount of time to conduct. In the embodiments of the present invention, the size of the preset delay window is set between 60 seconds and 180 seconds, depending on the type of seedling being transported, such as 120 seconds. In other embodiments of the present invention, the size of the preset delay window can be set according to specific circumstances, and will not be limited or elaborated here.

[0036] Step S3: Based on the average oxygen consumption rate, average vibration spectrum vector, initial frequency band sensitive weight vector, and corresponding initial covariance matrix of different sliding windows, obtain the frequency band sensitive weight optimization vector for each sliding window; based on the vibration characteristics and time distribution characteristics of the average vibration spectrum vector of different sliding windows, as well as the distribution of average oxygen consumption rate, obtain the oxygen consumption recovery time.

[0037] The average oxygen consumption rate reflects the total metabolic level of the organism within the sliding window. An increase in the average oxygen consumption rate indicates that the frequency combination corresponding to the vibration spectrum has triggered a strong stress response. The average vibration spectrum vector reflects the frequency domain energy distribution of water vibration within the sliding window. The initial frequency band sensitive weight vector reflects the prior estimate at the start of the algorithm. The initial covariance matrix reflects the uncertainty of the initial weight estimate. Therefore, based on the average oxygen consumption rate, average vibration spectrum vector, initial frequency band sensitive weight vector, and corresponding initial covariance matrix of different sliding windows, the frequency band sensitive weight optimization vector for each sliding window is obtained.

[0038] Preferably, in one embodiment of the present invention, the method for obtaining the frequency band sensitive weight optimization vector is described in [reference needed]. Figure 3 It illustrates a flowchart of a method for obtaining a frequency band sensitive weight optimization vector, including: Step S301: Based on the initial covariance matrix and the average vibration spectrum vector of the next sliding window, obtain the Kalman gain of the next sliding window.

[0039] Preferably, in one embodiment of the present invention, the method for obtaining the Kalman gain includes: The product between the initial covariance matrix and the average vibration spectrum vector of the next sliding window is obtained as the first product vector. The product between the transpose of the average vibration spectrum vector of the next sliding window and the first product vector is obtained as the second product value; the sum of the first product vector divided by the second product vector, the preset forgetting factor, and the preset regularization factor is obtained as the Kalman gain of the next sliding window.

[0040] The formula is expressed as: ; Indicates the next sliding window Kalman gain; Represents a sliding window The initial covariance matrix; Indicates the next sliding window The average vibration spectrum vector; Indicates a preset forgetting factor; This indicates the preset regularization factor; This represents the transpose symbol.

[0041] It should be noted that the physical condition of aquatic seedlings changes dynamically over time during transportation. To balance the algorithm's memory length and tracking ability, a forgetting factor is added. A smaller forgetting factor results in faster forgetting, making it overly sensitive to instantaneous noise and causing drastic parameter fluctuations; a larger forgetting factor leads to a slower response. In one embodiment of this invention, the preset forgetting factor is set within the range of 0.95-0.99, and can take any value within this range. To avoid the denominator potentially being 0, the preset regularization factor is set to a very small positive number, and its value can be specifically set according to the range of the denominator, such as... In other embodiments of the present invention, the magnitudes of the preset forgetting factor and the preset regularization factor can be set according to specific circumstances, and are not limited or elaborated here.

[0042] Step S302: Obtain the dot product between the initial frequency band sensitive weight vector and the average vibration spectrum vector of the next sliding window, and obtain the sum of the standard oxygen consumption rate and the dot product result as the predicted oxygen consumption rate of the next sliding window; based on the initial frequency band sensitive weight vector, the initial covariance matrix, the Kalman gain of the next sliding window, the average vibration spectrum vector, and the difference between the average oxygen consumption rate and the predicted oxygen consumption rate, obtain the frequency band sensitive weight optimization vector and the covariance matrix of the next sliding window.

[0043] It should be noted that, in the embodiments of the present invention, the method for obtaining the standard oxygen consumption rate is as follows: when the vehicle is stationary, the historical oxygen consumption rate is calculated as the standard oxygen consumption rate.

[0044] It should be noted that the dot product between vectors is the product of the transpose of one vector and another vector, that is, the product of the corresponding components of the two vectors and then the sum of the products.

[0045] It should be noted that, in the embodiments of the present invention, before any measurement data is available, in order to avoid deviation, the initial frequency band sensitive weight vector is set to an all-zero vector; the covariance matrix reflects the uncertainty of the initial frequency band sensitive weight vector, and the greater the initial uncertainty, the higher the initial covariance matrix should be. ,in, Represents a maximum positive number, such as This indicates a great uncertainty about the weights; The identity matrix is ​​represented by the initial frequency band sensitive weight vector and the initial covariance matrix. In other embodiments of the present invention, the initial frequency band sensitive weight vector and the initial covariance matrix may also be set according to specific circumstances, which are not limited or elaborated here.

[0046] Preferably, in one embodiment of the present invention, the method for obtaining the frequency band sensitive weight optimization vector and the covariance matrix includes: The product of the Kalman gain of the next sliding window and the difference between the average oxygen consumption rate and the predicted oxygen consumption rate is used as the frequency band adjustment vector; the sum of the frequency band adjustment vector and the initial frequency band sensitive weight vector is used as the frequency band sensitive weight optimization vector for the next sliding window. The formula is expressed as: ;in, Indicates the next sliding window Frequency band sensitive weight optimization vector; Represents a sliding window The initial frequency band sensitive weight vector; Indicates the next sliding window Kalman gain; Indicates the next sliding window The prior prediction error.

[0047] Based on this, the frequency band sensitive weight optimization vector can dynamically track changes in the seedling's physical condition. The smaller the weight component, the more the seedling gradually adapts to low-frequency shaking; the larger the weight component, the more the seedling becomes sensitive to high-frequency vibration due to fatigue.

[0048] Obtain the product vector between the Kalman gain of the next sliding window and the transpose of the average vibration spectrum vector; obtain the difference between the identity matrix and the product vector, calculate the product of the difference result and the initial covariance matrix, and add a preset forgetting factor to obtain the covariance matrix of the next sliding window.

[0049] The formula is expressed as: ;in, Indicates the next sliding window The covariance matrix; Indicates a preset forgetting factor; Represents the identity matrix; Indicates the next sliding window Kalman gain; Indicates the next sliding window The average vibration spectrum vector; Indicates the transpose symbol; Represents a sliding window The initial covariance matrix.

[0050] Step S303: Using the next sliding window as a new initial step, iteratively obtain the frequency band sensitive weight optimization vector and covariance matrix of the next sliding window to obtain the frequency band sensitive weight optimization vector for each sliding window.

[0051] It should be noted that, for the first sliding window, the optimized frequency-sensitive weight vector and covariance matrix of the first sliding window are obtained based on the initial frequency-sensitive weight vector, the initial covariance matrix, the Kalman gain of the first sliding window, the average vibration spectrum vector, and the difference between the average oxygen consumption rate and the predicted oxygen consumption rate. Then, the optimized frequency-sensitive weight vector of the first sliding window is used as the new initial frequency-sensitive weight vector, and the covariance matrix is ​​used as the new initial covariance matrix to obtain the second sliding window. This process is repeated iteratively to obtain the optimized frequency-sensitive weight vector for each sliding window.

[0052] In theory, after the external stimulus disappears, the oxygen consumption rate will decrease exponentially back to the baseline value, requiring analysis of the recovery status. Each element in the stationary vibration spectrum vector reflects the vehicle's energy transfer rate to each frequency band, quantifying the total vibration energy intensity across the entire frequency band. The greater the intensity, the less stable the vehicle is, and the longer the oxygen consumption recovery time. The temporal distribution characteristics reflect the decay trend, which helps quantify the oxygen consumption recovery status. Based on the vibration characteristics and temporal distribution characteristics of the average vibration spectrum vector of different sliding windows, as well as the average oxygen consumption rate distribution, the oxygen consumption recovery time can be obtained.

[0053] Preferably, in one embodiment of the present invention, the method for obtaining the oxygen recovery time includes: First, if the magnitude of the average vibration spectrum vector of the sliding window is less than the preset vibration threshold, the corresponding sliding window is taken as a stable sliding window; if the length of the time period formed by the continuous stable sliding window is greater than the preset duration threshold, the difference between the average oxygen consumption rate at the beginning of the time period formed by the continuous stable sliding window and the standard oxygen consumption rate is obtained as the relative difference. It should be noted that the magnitude of the average vibration spectrum vector reflects the total vibration energy intensity. The larger the magnitude, the less stable the vehicle operation; the smaller the magnitude, the smaller the vibration and the longer the duration, the greater the possibility of stable vehicle operation. The relative difference reflects the initial intensity of the stress response.

[0054] It should be noted that, in one embodiment of the present invention, the preset vibration threshold is set to the modulus of the average vibration spectrum vector of the water body when the vehicle is traveling at a constant speed on a standard smooth road surface; in order to ensure the reliability of the analysis, the longer the range of conditions that need to be met, the preset duration threshold is set to 10 minutes; in other embodiments of the present invention, the size of the preset vibration threshold and the preset duration threshold can be set according to the specific situation, and will not be limited or elaborated here.

[0055] The second step is to obtain the length of the continuous and stable sliding window period divided by the preset oxygen consumption recovery time and perform a negative correlation mapping as the time decay coefficient; obtain the product of the relative difference and the time decay coefficient, calculate the sum of the product result and the standard oxygen consumption rate, and use it as the oxygen consumption index recovery function. It should be noted that the larger the time decay coefficient, the longer the preset oxygen consumption recovery time, the slower the oxygen consumption rate decreases, and the longer it takes to return to the standard oxygen consumption rate.

[0056] It should be noted that, in the embodiments of the present invention, an exponential function with a base of the natural constant is used. Negative correlation mapping can be performed; in other embodiments of the present invention, negative correlation can also be performed by calculating the reciprocal; the specific means are well known to those skilled in the art and will not be described in detail here.

[0057] The third step is to obtain the mean of the average oxygen consumption rate of the continuous and stable sliding window as the overall oxygen consumption rate; and to obtain the oxygen consumption recovery time when the oxygen consumption exponential recovery function and the overall oxygen consumption rate are equal based on the nonlinear least squares method.

[0058] It should be noted that, based on the equation model formed by the oxygen consumption index recovery function and the overall oxygen consumption rate being equal, the nonlinear least squares method is used to fit the oxygen consumption recovery time. The specific nonlinear least squares method is a technique well known to those skilled in the art and will not be elaborated here.

[0059] Step S4: Based on the preset vehicle suspension transfer coefficients for different preset frequency bands, the standard pavement power spectral density of each candidate road segment, and the frequency band sensitive weight optimization vector of each sliding window, obtain the steady-state oxygen consumption ratio of each candidate road segment; based on the preset vehicle travel time, steady-state oxygen consumption rate, and oxygen consumption recovery time of each candidate road segment, obtain the comprehensive road segment cost of each candidate road segment.

[0060] The standard road surface spectrum only represents the unevenness of the ground and does not take into account the filtering and attenuation effects of the vehicle chassis suspension system and water tank structure on vibration. Directly using the road surface spectrum will exaggerate the impact of high-frequency vibration. By combining the vehicle suspension transmission coefficient and frequency band sensitive weight optimization vector analysis, the stimulation intensity of road segment on biological metabolism is quantified. Based on the preset vehicle suspension transmission coefficient of different preset frequency bands, the standard road surface power spectral density of each candidate road segment, and the frequency band sensitive weight optimization vector of each sliding window, the steady-state oxygen consumption ratio of each candidate road segment is obtained.

[0061] Preferably, in one embodiment of the present invention, the method for obtaining the steady-state oxygen consumption ratio of a road segment includes: Based on the preset vehicle suspension transmission coefficients for different preset frequency bands and the standard pavement power spectral density for each candidate road segment, the frequency band water vibration spectral vector of each candidate road segment is obtained. It should be noted that, in the embodiments of the present invention, the standard pavement power spectral density can be based on the ISO 8608 pavement roughness standard published by the International Organization for Standardization, and the standard pavement power spectral density corresponding to the road grade of the candidate road segment can be found.

[0062] Preferably, in one embodiment of the present invention, the method for obtaining the frequency band water body vibration spectrum vector includes: The product of the preset vehicle suspension transmission coefficient for different preset frequency bands and the standard pavement power spectral density for each candidate road segment is obtained as the water vibration energy of each candidate road segment in each preset frequency band. The column vector of water vibration energy in all preset frequency bands for each candidate road segment is formed according to the frequency band from smallest to largest, and is used as the frequency band water vibration spectrum vector.

[0063] The dot product between the frequency band sensitive weight optimization vector of each sliding window and the frequency band water vibration spectrum vector of each candidate road segment is obtained as the additional oxygen consumption increment. The ratio of the additional oxygen consumption increment to the standard oxygen consumption rate is obtained, and the sum of the ratio and the positive integer 1 is used as the steady-state oxygen consumption ratio of the road segment.

[0064] The formula is expressed as: ;in, Indicates the first Steady-state oxygen consumption ratio of each candidate road segment; Indicates the first Frequency-sensitive weight optimization vector for each sliding window; Indicates the first The frequency band water vibration spectrum vector of each candidate road segment; This represents the dot product between the frequency band sensitive weight optimization vector and the frequency band water body vibration spectrum vector, i.e., the additional oxygen consumption increment; This indicates the standard oxygen consumption rate.

[0065] The greater the additional oxygen consumption, the greater the steady-state oxygen consumption ratio of the road section.

[0066] The preset vehicle passage time reflects physical traffic efficiency; the longer the passage time, the slower the efficiency and the greater the loss cost. The steady-state oxygen consumption rate of the road segment reflects the real-time stimulation intensity of the road segment's bumpiness on organisms; the higher the steady-state oxygen consumption rate of the road segment, the more oxygen is consumed, and the higher the survival rate risk. The oxygen consumption recovery time reflects the difficulty of the seedlings' physiological function recovery; the longer the oxygen consumption recovery time, the greater the difficulty of recovery. Comprehensive analysis more fully quantifies the loss cost of candidate road segments. Based on the preset vehicle passage time, steady-state oxygen consumption rate, and oxygen consumption recovery time of each candidate road segment, the comprehensive road segment cost of each candidate road segment is obtained.

[0067] Preferably, the longer the preset vehicle travel time, the greater the steady-state oxygen consumption rate of the road segment, and the higher the metabolic accumulation during travel; the longer the oxygen consumption recovery time, the greater the steady-state oxygen consumption rate of the road segment, the greater the multiple of the additional oxygen consumption increment, the greater the additional physiological recovery time required, and the greater the overall road segment cost; in one embodiment of the present invention, the preset vehicle travel time, the steady-state oxygen consumption rate of the road segment, and the oxygen consumption recovery time are all positively correlated with the overall road segment cost.

[0068] In one embodiment of the present invention, the product of the preset vehicle travel time and the steady-state oxygen consumption rate of each candidate road segment is obtained as a first product; the difference between the steady-state oxygen consumption ratio of the road segment and the positive integer 1 is obtained to reflect the additional oxygen consumption ratio; the product of the oxygen recovery time of each candidate road segment and the difference result is obtained as a second product; the sum of the first product and the second product is obtained as the cost of biological metabolic loss. The formula is expressed as: ;in, Indicates the first The biological metabolic cost of each candidate road segment; Indicates the first The preset vehicle travel time for each candidate road segment; Indicates the first Steady-state oxygen consumption rate of each candidate road segment; Indicates the first Oxygen recovery time for each candidate road segment.

[0069] It should be noted that, in the embodiments of the present invention, the method for obtaining the preset vehicle travel time includes: using the average travel speed of historical vehicles in the candidate road segment as the standard travel speed; and calculating the length of the candidate road segment divided by the standard travel speed as the vehicle travel time.

[0070] The product of the preset balance coefficient and the vehicle passage time is obtained as the vehicle passage weighted duration; the difference between the positive integer 1 and the preset balance coefficient is obtained, and the product of the difference result and the biological metabolic loss cost is calculated as the biological metabolic loss weighted cost; the sum of the vehicle passage weighted duration and the biological metabolic loss weighted cost is obtained as the comprehensive road segment cost.

[0071] The formula is expressed as: ;in, Indicates the first The comprehensive road segment cost for each candidate road segment; This indicates the preset balance coefficient; Indicates the first The biological metabolic cost of each candidate road segment; Indicates the first The preset vehicle travel time for each candidate road segment.

[0072] It should be noted that, in one embodiment of the present invention, considering that the preset balance coefficient is used to adjust the system's emphasis on physical time, a larger preset balance coefficient results in the fastest path navigation, but also greater biological losses; if the preset balance coefficient is smaller, the system will take a serious detour and spend a long time. Considering that the survival rate of organisms is more important than the speed of time, in one embodiment of the present invention, the preset balance coefficient is set to 0.3. In other embodiments of the present invention, the size of the preset balance coefficient can be set according to specific circumstances, and will not be limited or elaborated here.

[0073] Step S5: Obtain the optimal transportation route based on the comprehensive road segment cost distribution of different candidate road segments.

[0074] The comprehensive route cost reflects the combined impact of candidate routes on transport survival rate and transport timeliness. The smaller the comprehensive route cost, the better it can balance the two states, which helps to plan the optimal transport route with a relatively slow transport timeliness and a high transport survival rate.

[0075] Preferably, in one embodiment of the present invention, the method for obtaining the optimal transportation route includes: Based on the comprehensive road segment cost of different candidate road segments, an optimal path search algorithm is performed on all candidate road segments to obtain the path with the minimum cumulative sum of the comprehensive road segment costs of all candidate road segments between the transportation origin and destination, which is taken as the optimal transportation path.

[0076] It should be noted that, in the embodiments of the present invention, the optimal path search algorithm may be Dijkstra's algorithm or A* algorithm; the specific means are well known to those skilled in the art and will not be described in detail here.

[0077] Based on this, in order to ensure that the navigation strategy always matches the real-time physiological state of the seedlings, the system sets a replanning cycle to recalculate the comprehensive road segment cost of the candidate road segments and then search for the latest optimal transportation route. If the newly searched optimal transportation route is inconsistent with the currently executed optimal transportation route, the system will generate new navigation instructions and push them to the driver's terminal interface, indicating that a change in the seedlings' sensitivity to the current road conditions has been detected and that a bio-friendly route has been optimized for you, ensuring that the transportation plan can adaptively cope with the evolution of the seedlings' state throughout their entire life cycle.

[0078] In summary, this invention analyzes the changing trends of vertical acceleration and dissolved oxygen concentration in water bodies at different times to obtain the frequency-sensitive weight optimization vector for each sliding window; based on the vibration characteristics and temporal distribution characteristics of the average vibration spectrum vector of different sliding windows, as well as the average oxygen consumption rate distribution, the oxygen recovery time is obtained; based on the preset vehicle suspension transfer coefficients of different preset frequency bands, the standard pavement power spectral density of each candidate road segment, and the frequency-sensitive weight optimization vector for each sliding window, the steady-state oxygen consumption ratio of each candidate road segment is obtained; combined with the preset vehicle travel time of each candidate road segment, the comprehensive road segment cost of each candidate road segment is obtained, leading to the optimal transportation route. This invention considers biological metabolism and traffic efficiency, accurately obtains the comprehensive road segment cost, and improves the effectiveness of route planning.

Claims

1. A method for planning high-survival-rate transportation routes for cross-regional aquatic seedlings, characterized in that, The method includes: Obtain the vertical acceleration of the water body and the dissolved oxygen concentration at every moment; A sliding window is constructed to traverse all time points. Based on the vertical acceleration of the water body at different time points, the power spectral density function of each sliding window is constructed. According to the distribution characteristics of the power spectral density function of each sliding window in different preset frequency bands and the trend of dissolved oxygen concentration at different time points within the corresponding preset delay window, the average vibration spectrum vector and average oxygen consumption rate of each sliding window are obtained. Based on the average oxygen consumption rate, average vibration spectrum vector, initial frequency band sensitive weight vector, and corresponding initial covariance matrix of different sliding windows, the frequency band sensitive weight optimization vector for each sliding window is obtained; based on the vibration characteristics and time distribution characteristics of the average vibration spectrum vector of different sliding windows, as well as the distribution of average oxygen consumption rate, the oxygen consumption recovery time is obtained. Based on the preset vehicle suspension transfer coefficients for different preset frequency bands, the standard pavement power spectral density for each candidate road segment, and the frequency band sensitivity weight optimization vector for each sliding window, the steady-state oxygen consumption ratio of each candidate road segment is obtained; based on the preset vehicle travel time, steady-state oxygen consumption rate, and oxygen consumption recovery time for each candidate road segment, the comprehensive road segment cost is obtained. The optimal transportation route is obtained based on the comprehensive road segment cost distribution of different candidate road segments.

2. The method for planning high survival rate transportation routes for cross-regional aquatic seedlings according to claim 1, characterized in that, The method for obtaining the power spectral density function includes: The power spectral density function is obtained by performing a Fourier transform on the sequence of vertical accelerations of the water body at different times within each sliding window.

3. The method for planning high survival rate transportation routes for cross-regional aquatic seedlings according to claim 1, characterized in that, The methods for obtaining the average vibration spectrum vector and the average oxygen consumption rate include: The integral value of the power spectral density function corresponding to each sliding window in each preset frequency band is obtained as the average vibration energy of each sliding window in each preset frequency band; The average vibration spectrum vector is formed by the column vectors of average vibration energy in different preset frequency bands from smallest to largest. Construct a straight line fitting the dissolved oxygen concentration at different times within a preset delay window for each sliding window, and obtain the absolute value of the slope of the straight line fitting the dissolved oxygen concentration as the average oxygen consumption rate.

4. The method for planning high survival rate transportation routes for cross-regional aquatic seedlings according to claim 1, characterized in that, The method for obtaining the frequency band sensitive weight optimization vector includes: Obtain the dot product between the initial frequency band sensitive weight vector and the average vibration spectrum vector of the next sliding window, and obtain the sum of the standard oxygen consumption rate and the dot product result as the predicted oxygen consumption rate for the next sliding window. Based on the initial covariance matrix and the average vibration spectrum vector of the next sliding window, the Kalman gain of the next sliding window is obtained. Based on the initial frequency band sensitive weight vector, the initial covariance matrix, the Kalman gain of the next sliding window, the average vibration spectrum vector, and the difference between the average oxygen consumption rate and the predicted oxygen consumption rate, the optimized frequency band sensitive weight vector and the covariance matrix of the next sliding window are obtained. Using the next sliding window as a new initial step, the frequency band sensitive weight optimization vector and covariance matrix of the next sliding window are obtained iteratively to obtain the frequency band sensitive weight optimization vector for each sliding window.

5. The method for planning a high survival rate transportation route for cross-regional aquatic seedlings according to claim 4, characterized in that, The methods for obtaining the frequency band sensitive weight optimization vector and the covariance matrix optimization vector include: The product of the Kalman gain of the next sliding window and the difference between the average oxygen consumption rate and the predicted oxygen consumption rate is used as the frequency band adjustment vector; the sum of the frequency band adjustment vector and the initial frequency band sensitive weight vector is used as the frequency band sensitive weight optimization vector for the next sliding window. Obtain the product vector between the Kalman gain of the next sliding window and the transpose of the average vibration spectrum vector; obtain the difference between the identity matrix and the product vector, calculate the product of the difference result and the initial covariance matrix, and add a preset forgetting factor to obtain the covariance matrix of the next sliding window.

6. The method for planning high survival rate transportation routes for cross-regional aquatic seedlings according to claim 1, characterized in that, The method for obtaining the oxygen recovery time includes: If the magnitude of the average vibration spectrum vector of a sliding window is less than the preset vibration threshold, the corresponding sliding window is taken as a stable sliding window; if the length of the period consisting of consecutive stable sliding windows is greater than the preset duration threshold, the difference between the average oxygen consumption rate at the beginning of the period consisting of consecutive stable sliding windows and the standard oxygen consumption rate is obtained as the relative difference. The length of the continuous and stable sliding window period is divided by the preset oxygen consumption recovery time, and a negative correlation mapping is performed to obtain the time decay coefficient; the product of the relative difference and the time decay coefficient is obtained, and the sum of the product result and the standard oxygen consumption rate is calculated to obtain the oxygen consumption exponential recovery function; the mean of the average oxygen consumption rate of the continuous and stable sliding window is obtained as the overall oxygen consumption rate. The oxygen consumption recovery time is obtained when the oxygen consumption index recovery function and the overall oxygen consumption rate are equal, based on the nonlinear least squares method.

7. The method for planning high survival rate transportation routes for cross-regional aquatic seedlings according to claim 1, characterized in that, The method for obtaining the steady-state oxygen consumption ratio of the road section includes: Based on the preset vehicle suspension transmission coefficients for different preset frequency bands and the standard pavement power spectral density for each candidate road segment, the frequency band water vibration spectral vector of each candidate road segment is obtained. The dot product between the frequency band sensitive weight optimization vector of each sliding window and the frequency band water vibration spectrum vector of each candidate road segment is obtained as the additional oxygen consumption increment. The ratio of the additional oxygen consumption increment to the standard oxygen consumption rate is obtained, and the sum of the ratio and the positive integer 1 is used as the steady-state oxygen consumption ratio of the road segment.

8. The method for planning high survival rate transportation routes for cross-regional aquatic seedlings according to claim 7, characterized in that, The method for obtaining the water vibration spectrum vector in the specified frequency band includes: The product of the preset vehicle suspension transmission coefficient for different preset frequency bands and the standard pavement power spectral density for each candidate road segment is obtained as the water vibration energy of each candidate road segment in each preset frequency band. The column vector of water vibration energy in all preset frequency bands for each candidate road segment is formed according to the frequency band from smallest to largest, and is used as the frequency band water vibration spectrum vector.

9. The method for planning high survival rate transportation routes for cross-regional aquatic seedlings according to claim 1, characterized in that, The preset vehicle travel time, steady-state oxygen consumption rate of the road segment, and oxygen consumption recovery time are all positively correlated with the comprehensive road segment cost.

10. The method for planning high survival rate transportation routes for cross-regional aquatic seedlings according to claim 1, characterized in that, The method for obtaining the optimal transportation route includes: Based on the comprehensive road segment cost of different candidate road segments, an optimal path search algorithm is performed on all candidate road segments to obtain the path with the minimum cumulative sum of the comprehensive road segment costs of all candidate road segments between the transportation origin and destination, which is taken as the optimal transportation path.

Images