Bend water flow optimization method based on rough strip flow guide effect evaluation index analysis

Abstract

The invention relates to the technical field of engineering hydraulics, in particular to a curve water flow optimization method based on rough strip flow guide effect evaluation index analysis, which comprises the following steps of: arranging a measurement section and a measurement point at a curve section to obtain water depth, flow velocity and flow data, introducing a water surface transverse gradient discrete coefficient and an average flow velocity discrete coefficient as evaluation indexes, and optimizing the curve water flow. The method comprises the following steps: determining a calculation section, constructing an original data matrix, carrying out forward and standardized processing on a coefficient, calculating an information bearing capacity, determining a weight, calculating a diversion speed reduction weighting coefficient by adopting a multi-index comprehensive weighting method, introducing a dynamic adjustment coefficient, and optimizing rough strip arrangement parameters in combination with a genetic algorithm and a local search strategy. The weighting coefficient adapts to the water flow energy change, the global optimal solution can be efficiently searched, the curve water flow form is improved, the downstream river channel scouring is relieved, and the stability of the release structure and the downstream river channel is improved.

Description

Technical Field

[0001] The invention relates to the technical field of engineering hydraulics, and in particular to a method for optimizing water flow in a curved channel based on the analysis of evaluation indexes of rough strip diversion effects. Background Art

[0002] Engineering hydraulics is an important technology, among which engineering hydraulics occupies a key position in the field of water conservancy engineering. It involves the study of the laws of water flow movement and its interaction with engineering structures, and provides important theoretical basis and technical support for the planning, design, construction and operation of various water conservancy projects.

[0003] In water conservancy engineering facilities, bend flow is a common and complex water flow phenomenon. Bend flow is affected by the centrifugal force and inertial force of the bend. The water flow in the trough presents an overall distribution pattern of "water accumulation on the concave bank and precipitation on the convex bank". The water surface has a large horizontal gradient. In addition, the water flow in the bend constantly adjusts its flow direction under the action of centrifugal force and inertial force. The energy of the water flow is redistributed in the bend, making the water flow at the outlet have higher kinetic energy, resulting in a large average flow velocity at the outlet section of the bend. The high-speed water flow carries huge energy to impact the downstream river channel, causing continuous uneven scouring of the riverbed and riverbank, which is not conducive to the long-term safe and stable operation of the discharge structure and the downstream natural river channel. In order to solve this technical problem, we provide a bend flow optimization method based on the analysis of rough strip diversion effect evaluation indicators. Summary of the invention

[0004] The object of the present invention is to provide a method for optimizing water flow in a curved channel based on the analysis of evaluation indicators of the diversion effect of rough strips, so as to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above objectives, one of the objectives of the present invention is to provide a method for optimizing water flow in a curved channel based on the analysis of rough strip diversion effect evaluation indicators, comprising the following steps:

[0006] S1. Arrange water depth measurement sections and flow velocity measurement sections along the curved section of the discharge structure, and arrange several measuring points from the concave bank to the convex bank of each section. Use water depth measuring instruments and flow velocity measuring instruments to measure, obtain water depth data at different positions of the curved section and flow velocity of different water flow layers, and finally use flow measuring instruments to measure and monitor the discharge flow of the discharge structure;

[0007] S2. Introducing the water surface horizontal gradient dispersion coefficient as the rough strip diversion effect evaluation index, and analyzing the water flow characteristics of the bend section, selecting two specific sections of the bend section of the discharge structure as the upstream section and the downstream section for calculating the coefficient, and determining the reference surface for calculation, and then introducing the average flow velocity dispersion coefficient as the rough strip speed reduction effect evaluation index, and according to the characteristics of the water flow velocity change in the bend, selecting two specific sections of the bend as the upstream section and the downstream section for calculating the coefficient;

[0008] S3, taking the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient as the target variables, constructing the original data matrix, and performing positive and normalization processing on the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient, calculating the information carrying capacity of each indicator through the formula, and finally determining the weights of the two indicators in the comprehensive evaluation based on the calculated information carrying capacity;

[0009] S4. Combined with the weight calculation results of the water surface lateral gradient dispersion coefficient and the average flow velocity dispersion coefficient, the multi-index comprehensive weighted method is used to obtain the diversion speed reduction weighted coefficient, and according to the calculated diversion speed reduction weighted coefficient and the water surface lateral gradient dispersion coefficient and the average flow velocity dispersion coefficient, the layout parameters of the rough strips are adjusted and optimized until the diversion speed reduction weighted coefficient reaches the maximum value and the water surface lateral gradient dispersion coefficient and the average flow velocity dispersion coefficient are minimized.

[0010] As a further improvement of the technical solution, when selecting the upstream section and the downstream section for calculating the water surface lateral gradient dispersion coefficient, S2 is determined based on the study of the energy balance and turbulence characteristics of the bend water flow, and the specific steps are as follows:

[0011] The whole flow field of the water flow in the bend section is measured to obtain the three-dimensional distribution information of the water flow velocity, and the water flow energy at different positions is calculated through the energy balance equation to determine the area with the largest turbulence intensity;

[0012] According to the geometric shape of the bend and the water flow conditions, the bend water flow energy distribution model is established to analyze the energy transfer path and dissipation law in the bend;

[0013] Combining the energy distribution model and turbulence characteristics analysis results, the two sections with the largest energy dissipation and the largest turbulence intensity are selected as the upstream and downstream sections for calculating the discrete coefficient of water surface lateral gradient.

[0014] As a further improvement of the technical solution, the selection of the upstream section and the downstream section for calculating the average velocity dispersion coefficient in S2 is determined based on the conservation of water flow momentum and vorticity distribution in the bend, as follows:

[0015] The velocity of the water flow in the bend is measured using a velocity meter to obtain time-varying velocity data, and the momentum change of the water flow at different positions is calculated using the momentum conservation equation. The distribution of vorticity is determined using the vorticity calculation method.

[0016] Construct a coupled model of water flow momentum and vorticity in a bend to analyze the momentum transfer process in the bend and the influence of vorticity on flow velocity changes;

[0017] Combined with the analysis results of the coupling model and the vorticity distribution, the two sections with the largest velocity gradient and the largest vorticity value are selected as the upstream and downstream sections for calculating the average velocity dispersion coefficient.

[0018] As a further improvement of the technical solution, the dynamic grouping processing method is used in S3 to construct the original data matrix, as follows:

[0019] According to the actual flow characteristics of the bend water flow, the rough strip arrangement area is divided into a bend entrance section, a bend curvature change section and a bend exit section, and the flow characteristics are determined by analyzing the flow velocity distribution of different sections;

[0020] According to the relative height differences of rough strips in different sub-areas, the measured water surface lateral gradient dispersion coefficient and average flow velocity dispersion coefficient data are grouped, and for each group, the change rate of the internal data is calculated as a dynamic change feature;

[0021] According to the dynamic change characteristics, a weight coefficient is assigned to each group, and the weighted group data are arranged and combined to construct the original data matrix.

[0022] As a further improvement of the technical solution, the forward and normalized adaptive processing algorithm for the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient in S3 is as follows:

[0023] The trend decomposition method is used to decompose the coefficient data into trend terms, period terms and random terms, which are used to analyze the time series trend of the measurement data;

[0024] For trend items, determine the direction and magnitude of positive conversion based on their growth or decline trend;

[0025] Calculate the period length and fluctuation amplitude of the periodic term and scale the periodic term accordingly;

[0026] By calculating the probability distribution characteristics of random items, a linear change algorithm is used to adjust their distribution to an approximate normal distribution;

[0027] According to the overall fluctuation range and distribution pattern of the data, the standardized scaling factor is dynamically determined, and the data is standardized according to the scaling factor.

[0028] As a further improvement of the technical solution, the modified calculation process for calculating the information carrying capacity of each indicator in S3 is as follows:

[0029] Based on the measured water depth and flow velocity data, the kinetic energy, potential energy and energy loss of the bend water flow at different positions are calculated. Based on the energy conversion, the influence of rough strips on the water flow energy conversion process is analyzed, and the quantitative relationship between the energy conversion efficiency and the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient is determined. Based on the quantitative relationship, the coefficient data is energy corrected.

[0030] Calculate the momentum components of the water flow in different directions in the bend, analyze the changes in the path and size of the water flow momentum transfer caused by the rough strip, establish a momentum transfer model to obtain the correlation between the momentum transfer characteristic parameters and the coefficients, and perform momentum correction on the coefficients based on the correlation;

[0031] The coefficient data after energy correction and momentum correction are substituted into the comparative and contradictory formulas for calculating information carrying capacity. At the same time, the Pearson correlation coefficient with energy and momentum weighting is adjusted to calculate the information carrying capacity. Finally, the weights of the two indicators in the comprehensive evaluation are determined based on the calculated information carrying capacity.

[0032] As a further improvement of the technical solution, when the multi-index comprehensive weighted method is used in S4 to calculate the diversion speed reduction weighted coefficient, a dynamic adjustment coefficient based on water flow energy balance is introduced. The specific steps are as follows:

[0033] Calculate the kinetic energy and potential energy of the bend water flow at different positions and the amount of energy dissipated by the rough strips, establish the water flow energy balance equation, and analyze the coupling degree between the water surface horizontal gradient discrete coefficient and the average flow velocity discrete coefficient and the water flow energy based on the energy balance relationship to determine the energy influencing factor;

[0034] A dynamic adjustment coefficient model is constructed based on the energy influencing factor. When the uneven distribution of water flow energy exceeds the preset threshold, the adjustment coefficient of the water surface horizontal gradient discrete coefficient is increased according to the preset proportion, and the dynamic adjustment coefficient of the average flow velocity discrete coefficient is adjusted according to the same proportion.

[0035] The dynamic adjustment coefficient is combined with the fixed weighting coefficient based on the weight calculation result to obtain the corrected comprehensive weighting coefficient, and the diversion and speed reduction weighting coefficient is calculated.

[0036] As a further improvement of the technical solution, when adjusting and optimizing the rough stripe arrangement parameters in S4, an optimization method based on a combination of a genetic algorithm and a local search strategy is adopted, as follows:

[0037] The relative height of the rough strips is encoded to form chromosome individuals, and the population is initialized. The weighted coefficient of the diversion speed reduction is used as the fitness function to calculate the fitness value of each individual in the population.

[0038] The selection, crossover and mutation operations of the genetic algorithm are used to iteratively evolve the population and find the global optimal solution area. In each generation of evolution, the individuals with the highest fitness are retained, the excellent genes between individuals are exchanged through the crossover operation, and new genes are introduced through the mutation operation.

[0039] When the genetic algorithm iterates to the preset generation, the gradient descent method is used to search the current optimal solution area. In the local search process, the relative height of the rough strips is optimized with the diversion speed reduction weighting coefficient, the water surface horizontal gradient discrete coefficient and the average flow velocity discrete coefficient as the target, until the optimal rough strip relative height combination is found that maximizes the diversion speed reduction weighting coefficient and minimizes the water surface horizontal gradient discrete coefficient and the average flow velocity discrete coefficient.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] In the bend water flow optimization method based on the analysis of the evaluation index of the rough strip diversion effect, the diversion speed reduction weighting coefficient is calculated by the multi-index comprehensive weighted method, and the dynamic adjustment coefficient based on the water flow energy balance is introduced in the calculation process, so that the weighting coefficient can adapt to the energy change of the water flow in real time, and better guide the optimization of the rough strip layout parameters. The optimization method combining the genetic algorithm with the local search strategy can efficiently find the global optimal solution and quickly converge to the best combination of rough strip layout parameters to maximize the diversion speed reduction weighting coefficient. At the same time, the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient are minimized, which helps to significantly improve the bend water flow morphology, reduce the average flow velocity of the outlet section, reduce the uneven scouring of the downstream river channel, and improve the long-term safety and stability of the discharge structure and the downstream river channel. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is the overall workflow diagram of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] Rough strip energy dissipation works are simple in shape and easy to construct. In recent years, research on rough strips has been favored by scholars, but there is a lack of research and publication of relevant patent technologies. Combined with existing research reports, it is found that scholars have focused on the basic hydraulic characteristics of rough strips, and the evaluation indicators are general, repetitive, and diverse. This makes it impossible to take into account the advantages and disadvantages of many indicators at the same time during engineering design, and the engineering reference and universality are not strong.

[0045] The present invention proposes to organically unify the surface gradient dispersion coefficient and the average flow velocity dispersion coefficient of the rough strip to the curved water flow into a diversion and speed reduction weighted coefficient. During engineering design, whether the rough strip arrangement pattern is reasonable can be judged only by the size of this evaluation index, which greatly improves the engineering applicability of the rough strip and enriches the evaluation method of the advantages and disadvantages of the rough strip arrangement pattern.

[0046] The hydraulic characteristics of water flow in a bend are complex, and there are significant differences in water depth, flow velocity and flow rate at different locations and water flow layers. In order to fully and accurately understand the movement state of water flow in a bend, it is necessary to measure at multiple locations and different water flow layers. By arranging multiple measurement sections along the way and arranging multiple measuring points on the sections, detailed water flow information can be obtained, providing sufficient data basis for subsequent analysis and evaluation.

[0047] See also Figure 1 As shown, this embodiment provides a method for optimizing water flow in a curved channel based on the analysis of rough strip diversion effect evaluation indexes, comprising the following steps:

[0048] S1. Determine the number of depth measurement sections and velocity measurement sections based on the length of the bend, the change in curvature, and the estimated complexity of the water flow characteristics. Intensify the arrangement of measurement sections near the starting point, midpoint, and end point of the bend to gain a deeper understanding of the three-dimensional flow characteristics of the bend water flow. For a bend section with a length of 50 meters and a gradually changing radius of curvature, arrange 8-10 measurement sections along the way to ensure that the changes in the water flow at different positions in the bend can be captured.

[0049] On each measuring section, measuring points are arranged at equal intervals from the concave bank to the convex bank or at unequal intervals according to the changing rules of water flow characteristics. In the areas near the concave bank and convex bank, the measuring points should be appropriately increased because the water flow is greatly affected by the boundary. In the area with a large water flow velocity gradient near the concave bank, the measuring point spacing can be set to 0.2-0.3 meters. In the relatively uniform area in the middle of the bend, the measuring point spacing can be increased to 0.5-0.8 meters. The number of measuring points is determined according to the section width and measurement accuracy requirements. 5-10 measuring points are arranged on each section to obtain sufficient information to accurately describe the water depth and flow velocity distribution in the section.

[0050] During measurement, the sensor of the depth measuring instrument is placed vertically at a certain height above the water surface to ensure that the ultrasonic wave emitted by the sensor can vertically penetrate the water surface and reflect back. The measuring instrument measures the time difference between ultrasonic emission and reception, combined with the propagation speed of sound waves in water, to calculate the water depth at the measuring point, effectively reducing measurement errors and improving data accuracy.

[0051] The flow meter is installed on a special measuring bracket so that its probe can be accurately aligned with the direction of water flow. During measurement, the flow meter emits sound wave signals and calculates the flow velocity by analyzing the frequency shift of the sound waves on the scatterers in the water flow, effectively reducing measurement errors and improving data accuracy.

[0052] The flow meter uses the principle of electromagnetic induction. When water flows through a magnetic field, the flow rate is calculated based on the relationship between the induced electromotive force and the flow velocity, effectively reducing measurement errors and improving data accuracy.

[0053] S2. Introducing the water surface horizontal gradient dispersion coefficient as the rough strip diversion effect evaluation index, and analyzing the water flow characteristics of the bend section, selecting two specific sections of the bend section of the discharge structure as the upstream section and the downstream section for calculating the coefficient, and determining the reference surface for calculation, and then introducing the average flow velocity dispersion coefficient as the rough strip speed reduction effect evaluation index, and according to the characteristics of the water flow velocity change in the bend, selecting two specific sections of the bend as the upstream section and the downstream section for calculating the coefficient;

[0054] When selecting the upstream and downstream sections for calculating the water surface gradient dispersion coefficient, the specific steps are as follows based on the study of the energy balance and turbulence characteristics of the bend flow:

[0055] The flow field is measured by a flow meter to obtain the three-dimensional velocity data (u, v, w) of the bend water flow at different positions, where u, v, and w represent the components of the velocity in the x, y, and z directions, respectively, and x is the mainstream direction of the bend. Based on the measured velocity data, the energy balance equation is used to calculate the water flow energy. The expression of the energy balance equation is:

[0056]

[0057] Where E is the total energy per unit mass of water, ρ is the density of water, g is the acceleration of gravity, and h is the height of the measuring point relative to the reference plane. During the calculation process, the energy values ​​of the measuring points at different depths of each measuring position are calculated respectively, so as to obtain the spatial distribution of the water flow energy of the entire bend section. The three-dimensional flow velocity data can fully describe the movement of the water flow. The energy balance equation takes into account the kinetic energy and potential energy. The calculated energy distribution fully reflects the energy state of the water flow, which helps to deeply understand the energy characteristics of the bend water flow and provide a basis for analyzing the laws of energy transfer and dissipation.

[0058] For three-dimensional flow velocity (u, v, w), the turbulence intensity calculation formula is: where σ u , σ v , σ wThey are the pulsation standard deviations of the flow velocity in the u, v, and w directions respectively. By calculating the turbulence intensity of each measuring point in the entire bend area, the area with the largest turbulence intensity is determined, which is helpful for in-depth research on the turbulence characteristics of the bend water flow and provides a basis for combined analysis with energy distribution, which is crucial to a comprehensive understanding of the bend water flow characteristics.

[0059] The energy distribution model of the bend water flow is established according to the geometric shape of the bend and the water flow conditions. The energy distribution model expression is: When establishing the model, it is necessary to determine the boundary conditions according to the actual situation. The inlet boundary conditions are set as the velocity distribution and water level height, and the outlet boundary conditions are set as the given pressure. Reasonable determination of the boundary conditions can improve the fit between the model and the actual situation, provide a powerful tool for analyzing the energy transfer path and dissipation law, and help optimize engineering applications such as rough strip layout. The above equations are solved using numerical calculation methods to obtain the velocity field and pressure field of the bend water flow, and then the energy distribution is calculated to provide a reliable basis for subsequent analysis and enhance the credibility of the research results. The model is used to analyze the energy transfer path in the bend, and observe the energy aggregation, transfer and diffusion in different areas of the bend. At the same time, the energy dissipation law is analyzed to determine in which areas the energy is mainly lost due to friction, vortex and other factors. The energy dissipation rate of different areas is calculated according to the energy distribution model, which helps to optimize the bend water flow control measures.

[0060] Combining the maximum energy dissipation area obtained by the energy distribution model analysis and the maximum turbulence intensity area determined by actual measurement, two sections are selected as the upstream section and downstream section for calculating the discrete coefficient of the water surface gradient. d (x) is the energy dissipation value of the section at position x, T(x) is the turbulence intensity value of the section at position x, and by analyzing E d The distribution curves of (x) and T(x) in the curve, select E d The two positions x1 and x2 where (x) and T(x) are simultaneously the largest are taken as the positions of the upstream section and the downstream section. The section is selected by comprehensively considering energy dissipation and turbulence intensity, which can accurately reflect the diversion effect of the rough strips, help optimize the layout of the rough strips, and improve the water flow control effect of the bend.

[0061] The selection of the upstream and downstream sections for the calculation of the average velocity dispersion coefficient in S2 is determined based on the conservation of water flow momentum and vorticity distribution in the bend, as follows:

[0062] According to the data measured by the flow meter, the momentum conservation equation is used to calculate the change of water flow momentum at different positions. The equation expression is: Where p is the pressure and μ is the dynamic viscosity coefficient. For each measuring point, the momentum change of the point is calculated by measuring the flow velocities u and v and the known water density ρ and dynamic viscosity coefficient μ, combined with the position information of the measuring point. The partial derivative calculation is effectively processed to improve the calculation accuracy, which helps to deeply understand the momentum transfer characteristics of water flow in bends and provide key information for analyzing the movement laws of water flow and the effect of rough strips.

[0063] The vorticity calculation method is used to determine the distribution of vorticity. The calculation formula of vorticity ω is: For each measuring point, the vorticity value of the point is calculated based on the measured flow velocity u and v data. By calculating the vorticity at each measuring point in the bend, a vorticity contour map or cloud map is drawn to intuitively display the distribution of vorticity in the bend, determine the area with larger vorticity, and study the influence of vorticity on flow velocity and provide a basis for selecting calculation sections.

[0064] When constructing the coupled model of water flow momentum and vorticity in a bend, the momentum conservation equation and the related equations of vorticity are used as the basis, and the model is constructed by considering the interaction between vorticity and momentum. This can comprehensively describe the complex characteristics of water flow in a bend, help reveal the intrinsic connection between momentum transfer and vorticity influence, and improve the level of understanding of water flow in a bend. The coupled model is solved by numerical calculation methods, and the bend area is divided into multiple small calculation units. The model equation is discretized on each unit, and then the flow velocity, vorticity and other variables in each unit are solved through iterative calculation. The results obtained from the solution are used to analyze the momentum transfer process in the bend, observe how momentum is transferred from the entrance to the exit of the bend, and the changes in the transfer direction and size in different areas of the bend. This helps to deeply understand the momentum transfer law and vorticity influence mechanism of water flow in the bend, and provide a scientific basis for engineering design and water flow optimization.

[0065] Combined with the momentum transfer characteristics and vorticity distribution obtained by coupling model analysis, the two sections with the largest velocity gradient and the largest vorticity value are selected as the upstream and downstream sections for calculating the average velocity dispersion coefficient, as follows:

[0066] Let G(x) be the comprehensive value of the velocity gradient at the section at position x, ω(x) is the vorticity value of the section at position x. By analyzing the distribution curves of G(x) and ω(x) in the bend, the two positions x3 and x4 where G(x) and ω(x) are simultaneously the largest are selected as the positions of the upstream and downstream sections. The sections selected in this way can better reflect the influence of vorticity on flow velocity changes and the key characteristics of momentum transfer, so that the calculated average flow velocity dispersion coefficient can more accurately evaluate the speed-reducing effect of the rough strip on the water flow in the bend.

[0067] S3, taking the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient as the target variables, constructing the original data matrix, and performing positive and normalization processing on the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient, calculating the information carrying capacity of each indicator through the formula, and finally determining the weights of the two indicators in the comprehensive evaluation based on the calculated information carrying capacity;

[0068] In S3, the dynamic grouping method is used to construct the original data matrix, as follows:

[0069] Multiple measuring sections are set up in the bend water flow, and the flow velocity at different positions on each section is measured using a flow velocity meter. Then the flow velocity distribution of different sections is analyzed. According to the flow velocity distribution analysis results, the rough strip layout area is divided into the bend entrance section, the bend curvature change section and the bend exit section. This helps to gain a deeper understanding of the velocity characteristics of the bend water flow, so as to better study the influence of rough strips at different positions on the bend water flow.

[0070] The measured data of the dispersion coefficient of the water surface transverse gradient and the dispersion coefficient of the average flow velocity are grouped according to the differences in the relative heights of the rough strips in different sub-areas. Assuming that there are n different situations of the relative heights of the rough strips, the data are divided into n groups. For each group of data, it includes the measured values ​​of the dispersion coefficient of the water surface transverse gradient and the dispersion coefficient of the average flow velocity at the bend entrance section, the bend curvature change section and the bend exit section at the corresponding relative height of the rough strip. This is convenient for studying the relationship between the relative height of the rough strip and the water flow characteristics, and provides an ordered data structure for subsequent calculations.

[0071] For each group of data, the change rate of its internal data is calculated as the dynamic change feature. Suppose a group of data is x1, x2, ..., x m , then the formula for calculating the rate of change is Then the average change rate of the group is calculated as the dynamic change feature of the group. The average change rate reflects the drastic degree of change of the group data under different rough strip arrangements, highlighting the differences in data characteristics of different groups, so that the weight distribution can better reflect the differences in data importance.

[0072] According to the calculated dynamic change characteristics, a weight coefficient is assigned to each group. Suppose the dynamic change characteristics of the jth group are The weight coefficient The larger the dynamic change characteristics of the group, the larger its weight coefficient, which means that the group data is more important in reflecting the impact of rough strips on the water flow in the bend. It can highlight the contribution of key group data to the overall evaluation and improve the accuracy of the data matrix for the assessment of the impact of rough strips.

[0073] The weighted grouped data are arranged and combined to construct the original data matrix. Assume that the number of groups is k and the number of evaluation indicators is p, where p = 2, namely the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient. Then the original data matrix X is a k×p matrix. For the jth group, the weighted water surface horizontal gradient dispersion coefficient is x' j1 =ω j × j1 ;x j1 is the unweighted water surface gradient dispersion coefficient, and the weighted average velocity dispersion coefficient is x' j2 =ω j × j2 ;x j2 is the unweighted average velocity dispersion coefficient. These weighted data are arranged in a matrix in sequence. This original data matrix can more reasonably reflect the characteristics of the bend flow under different rough strip arrangements, providing a more representative data basis for subsequent analysis and evaluation, and helping to improve the scientificity and reliability of the entire evaluation system, thereby more accurately evaluating the diversion and speed reduction effect of the rough strips and optimizing the rough strip arrangement plan.

[0074] In S3, an adaptive processing algorithm is used to normalize and standardize the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient, as follows:

[0075] Firstly, the exponential smoothing method is used to decompose the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient data into trend, cycle and randomness to analyze the time series trend of the measurement data.

[0076] For trend items, if there is an upward trend, it is converted to an increasing direction, otherwise it is converted to a decreasing direction. At the same time, the conversion amplitude is determined according to the trend slope. Then the cycle length and fluctuation amplitude of the periodic item are calculated, and the periodic item is scaled according to the fluctuation size. Then the probability distribution characteristics of the random item are calculated, and the linear change algorithm is used to make its distribution approximate to a normal distribution. Finally, based on the overall fluctuation range and distribution form of the data, the standardized scaling factor is dynamically calculated, and the data is standardized using this factor so that the data maintains the relative relationship of the original data characteristics within a specific interval, thereby improving the data comparability and stability, and providing a good foundation for subsequent analysis.

[0077] Water flow in a bend is affected by many complex physical factors, such as energy conversion and momentum change. By calculating the kinetic energy, potential energy and energy loss at different positions, we can fully understand the energy state of the water flow. Because the presence of rough strips will change the energy conversion process of the water flow, it is necessary to correct the coefficient based on the quantitative relationship between the energy conversion efficiency and the discrete coefficient of the water surface transverse gradient and the discrete coefficient of the average flow velocity.

[0078] The revised calculation process for calculating the information carrying capacity of each indicator in S3 is as follows:

[0079] Based on the measured water depth and flow velocity data, the kinetic energy E of the bend water flow at different positions is calculated. k Potential energy E p and energy loss ΔE. According to the energy conversion analysis, the influence of rough strips on the energy conversion process of water flow is determined, and the quantitative relationship between the energy conversion efficiency η and the discrete coefficient of water surface gradient CW and the discrete coefficient of average flow velocity CV is determined. It is assumed that the relationship obtained by theoretical analysis and experimental data fitting is η=aCW+bCV+c; a, b, c are all fitting coefficients. According to this relationship, the coefficient data is energy corrected. Suppose the discrete coefficient of water surface gradient before correction is CW0, and the discrete coefficient of average flow velocity is CV0. The corrected coefficients are CW1 and CV1 respectively, then CW1=CW0(1+αη); CV1=CV0(1+βη); α and β are correction coefficients determined according to the degree of energy influence. This calculation process comprehensively considers the energy and momentum change process of water flow, and can more realistically simulate the complex influence of rough strips on bend water flow.

[0080] Calculate the momentum components of the bend flow in different directions. Assume that the bend flow mainly moves in the horizontal plane. Let the x direction be the mainstream direction of the bend, the y direction be the lateral direction, and the components of the flow velocity in the x and y directions be u and v respectively. Then the momentum component in the x direction P is x =ρhu, the momentum component in the y direction is P y =ρhv; Analyze the changes of rough strips on the transfer path and size of water flow momentum, establish a momentum transfer model, and obtain the correlation between the momentum transfer characteristic parameters and the coefficients through model analysis, set as k1=dCW+eCV+f; k2=gCW+hCV+i; d, e, f, g, h, i are derivatives determined according to the model, and the coefficients are momentum corrected according to the correlation. Suppose the corrected water surface horizontal gradient discrete coefficient is CW2, and the average flow velocity discrete coefficient is CV2, then CW2=CW1(1+γk1); CV2=CV1(1+δk2); γ and δ are correction coefficients determined according to the degree of momentum influence.

[0081] Substitute the coefficient data CW2 and CV2 after energy correction and momentum correction into the formula for calculating the contrast and contradiction of information carrying capacity. The contrast is expressed by the standard deviation σ j The standard deviations of the water surface gradient dispersion coefficient CW2 and the average velocity dispersion coefficient CV2 are respectively The contradiction reflects the correlation between different indicators. Let the contradiction between indicators CW2 and CV2 be f CW2，CV2 , and adjust the Pearson correlation coefficient with energy and momentum weighting, and set the Pearson correlation coefficient to Assume the weighted correlation coefficient is The information carrying capacity of indicator CW2 is The information carrying capacity of indicator CV2 is

[0082] Finally, the weights of the two indicators in the comprehensive evaluation are determined based on the calculated information carrying capacity, and the weight of the water surface horizontal gradient dispersion coefficient is Weight of the coefficient of dispersion of the mean flow rate Through such a calculation process, the influence of energy and momentum factors on the indicators is comprehensively considered, making the weight determination more scientific and reasonable, and can more accurately reflect the importance of the diversion and deceleration effects of rough strips on the water flow in the bend in the comprehensive evaluation.

[0083] S4. Combined with the weight calculation results of the water surface lateral gradient dispersion coefficient and the average flow velocity dispersion coefficient, the multi-index comprehensive weighted method is used to obtain the diversion speed reduction weighted coefficient, and according to the calculated diversion speed reduction weighted coefficient and the water surface lateral gradient dispersion coefficient and the average flow velocity dispersion coefficient, the layout parameters of the rough strips are adjusted and optimized until the diversion speed reduction weighted coefficient reaches the maximum value and the water surface lateral gradient dispersion coefficient and the average flow velocity dispersion coefficient are minimized.

[0084] When S4 uses the multi-index comprehensive weighted method to calculate the diversion speed reduction weighted coefficient, a dynamic adjustment coefficient based on water flow energy balance is introduced. The specific steps are as follows:

[0085] According to the kinetic energy and potential energy of the bend water flow at different positions and the dissipation of water flow energy by the rough strip, the water flow energy balance equation is established, and the coupling degree between the water surface horizontal gradient discrete coefficient CW2 and the average flow velocity discrete coefficient CV2 and the water flow energy is analyzed according to the energy balance relationship to determine the energy influencing factor and

[0086] Based on the energy influencing factor, a dynamic adjustment coefficient model is constructed, and the dynamic adjustment coefficient of the water surface horizontal gradient discrete coefficient is set as The dynamic adjustment coefficient of the average flow velocity dispersion coefficient is The model is χ and ε are coefficients determined according to the degree of energy influence. When the uneven distribution of water flow energy exceeds the preset threshold, the adjustment coefficient of the water surface horizontal gradient discrete coefficient is increased according to the preset proportion, and the dynamic adjustment coefficient of the average flow velocity discrete coefficient is adjusted according to the same proportion. The threshold is set as T. When the calculated water flow energy uneven distribution index I>E, the increase ratio is set as p, then the new

[0087] Combining the dynamic adjustment coefficient with the fixed weighting coefficient based on the weight calculation result, the revised comprehensive weighting coefficient is W' CW =WCW×k' CW ;W' CV =WCV×k' CV ; Finally, calculate the diversion speed reduction weighted coefficient SS, the calculation formula is SS = W'CW ×CW+W' CV ×CV; by introducing a dynamic adjustment coefficient based on water flow energy balance, the coefficient weight can be adjusted in real time according to the actual situation of water flow energy, so that the diversion and speed reduction weighted coefficient can more accurately reflect the diversion and speed reduction effect of the rough strip under different water flow energy conditions, thereby improving the accuracy and adaptability of the evaluation.

[0088] The flow in a bend is affected by many factors, and its hydraulic characteristics are complex and changeable. The effect of rough strips on the flow in a bend not only involves the change of flow velocity and water surface morphology, but also is interrelated with factors such as water flow energy and momentum. Therefore, it is necessary to comprehensively consider these factors to accurately evaluate the diversion and deceleration effect of rough strips, which prompts the use of a multi-index comprehensive weighted method, the introduction of a dynamic adjustment coefficient based on water flow energy balance, and an optimization method combining genetic algorithms with local search strategies.

[0089] When adjusting and optimizing the rough stripe arrangement parameters in S4, an optimization method based on a combination of genetic algorithm and local search strategy is adopted, as follows:

[0090] Encode the relative height of the rough strip, and set the value range of the relative height h of the rough strip to [h min ,h max ], dividing it into m intervals, so that each encoded binary string represents a chromosome individual, which corresponds to a possible value of the relative height of the rough stripe, and randomly generates o chromosome individuals to form an initial population, avoiding the problem of falling into local optimality by searching from only a single or a few points. With the increase of the number of iterations, the relative height of the rough stripe represented by the individuals in the population will be closer and closer to the optimal value.

[0091] The diversion and speed reduction weighted coefficient SS is used as the fitness function. For each individual in the population, the corresponding water surface horizontal gradient dispersion coefficient CW2 and the average flow velocity dispersion coefficient CV2 are calculated according to the current relative height of the rough strips. Then, the diversion and speed reduction weighted coefficient SS is calculated according to the multi-index comprehensive weighted method. The larger the value of SS, the better the diversion and speed reduction effect of the rough strip arrangement corresponding to the individual, and the higher its fitness.

[0092] The roulette wheel selection method is used for selection, and the proportion of each individual's fitness value to the total fitness value is calculated as the probability of its being selected. Then, individuals are randomly selected based on these probabilities. Individuals with high fitness have a greater chance of being selected, thereby retaining better gene combinations for the next generation.

[0093] Select two parent individuals as A and B, randomly select a crossover point, then combine the first k bits of A and the last ok bits of B into a new individual C, and combine the first k bits of B and the last ok bits of A into a new individual D. Through the crossover operation, the exchange of excellent genes is achieved and the diversity of the population is increased.

[0094] For each individual, a gene position is randomly selected for mutation according to the preset mutation probability, which helps to jump out of the local optimal solution and explore a wider solution space. When the genetic algorithm iterates to the preset generation, the gradient descent method is used for local search. The objective function J is constructed with the diversion speed reduction weighted coefficient SS, the water surface horizontal gradient discrete coefficient CW2 and the average flow velocity discrete coefficient CV2 as the target, then J = -SS + κCW2 + λCV2; κ and λ are weight coefficients used to balance the importance of the three indicators. Here, in order to minimize CW2 and CV2, a positive value is taken in the objective function, and the gradient of the objective function J with respect to the relative height h of the rough strip is calculated.

[0095] According to the iterative formula of the gradient descent method Update the relative height of the rough strip, h new Relative height of rough strip after not updating, h old is the current relative height of the rough strips, θ is the learning rate, which is used to control the step size. In each iteration, the new values ​​of SS, CW2 and CV2 are calculated, and it is determined whether the stopping condition is met. The stopping condition is to reach the maximum number of iterations. This process is repeated continuously to optimize the relative height of the rough strips until the optimal relative height combination of the rough strips is found that maximizes the diversion and deceleration weighted coefficient and minimizes the water surface lateral gradient discrete coefficient and the average flow velocity discrete coefficient. The rough strip arrangement parameters corresponding to this optimal combination can achieve the best diversion and deceleration effect in the local range. Combined with the global search of the genetic algorithm in the early stage, the final result can be close to the global optimal solution in the entire solution space, effectively improving the rationality and effectiveness of the rough strip arrangement and improving the flow characteristics of the bend water flow.

[0096] The above-mentioned bend water flow optimization method based on the analysis of rough strip diversion effect evaluation indicators includes arranging measurement sections and measuring points in the bend section to obtain water depth, flow velocity and flow rate data, introducing the water surface horizontal gradient discrete coefficient and the average flow velocity discrete coefficient as evaluation indicators, determining the calculation section, constructing the original data matrix, performing forward and standardization processing on the coefficients, calculating the information carrying capacity and determining the weights, using a multi-indicator comprehensive weighted method to calculate the diversion speed reduction weighted coefficient, introducing a dynamic adjustment coefficient, and combining genetic algorithms with local search strategies to optimize the rough strip arrangement parameters. The present invention comprehensively considers multiple factors, makes the weighted coefficient adapt to the change of water flow energy, can efficiently find the global optimal solution, improves the bend water flow morphology, reduces the scouring of the downstream river channel, and improves the stability of the discharge structure and the downstream river channel.

[0097] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for optimizing water flow in a curved channel based on the evaluation index analysis of rough strip diversion effect, characterized in that: The following steps are involved: S1. Arrange water depth measurement sections and flow velocity measurement sections along the curved section of the discharge structure, and arrange several measuring points from the concave bank to the convex bank of each section. Use water depth measuring instruments and flow velocity measuring instruments to measure, obtain water depth data at different positions of the curved section and flow velocity of different water flow layers, and finally use flow measuring instruments to measure and monitor the discharge flow of the discharge structure; S2. Introducing the water surface horizontal gradient dispersion coefficient as the rough strip diversion effect evaluation index, and analyzing the water flow characteristics of the bend section, selecting two specific sections of the bend section of the discharge structure as the upstream section and the downstream section for calculating the coefficient, and determining the reference surface for calculation, and then introducing the average flow velocity dispersion coefficient as the rough strip speed reduction effect evaluation index, and according to the characteristics of the water flow velocity change in the bend, selecting two specific sections of the bend as the upstream section and the downstream section for calculating the coefficient; S3, taking the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient as the target variables, constructing the original data matrix, and performing positive and normalization processing on the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient, calculating the information carrying capacity of each indicator through the formula, and finally determining the weights of the two indicators in the comprehensive evaluation based on the calculated information carrying capacity; S4. Combined with the weight calculation results of the water surface lateral gradient dispersion coefficient and the average flow velocity dispersion coefficient, the multi-index comprehensive weighted method is used to obtain the diversion speed reduction weighted coefficient, and according to the calculated diversion speed reduction weighted coefficient and the water surface lateral gradient dispersion coefficient and the average flow velocity dispersion coefficient, the layout parameters of the rough strips are adjusted and optimized until the diversion speed reduction weighted coefficient reaches the maximum value and the water surface lateral gradient dispersion coefficient and the average flow velocity dispersion coefficient are minimized.

2. The method for optimizing water flow in a curved channel based on the analysis of rough strip diversion effect evaluation indexes according to claim 1, characterized in that: When selecting the upstream section and the downstream section for calculating the water surface lateral gradient dispersion coefficient, S2 is determined based on the study of the energy balance and turbulence characteristics of the bend water flow. The specific steps are as follows: The whole flow field of the water flow in the bend section is measured to obtain the three-dimensional distribution information of the water flow velocity, and the water flow energy at different positions is calculated through the energy balance equation to determine the area with the largest turbulence intensity; According to the geometric shape of the bend and the water flow conditions, the bend water flow energy distribution model is established to analyze the energy transfer path and dissipation law in the bend; Combining the energy distribution model and turbulence characteristics analysis results, the two sections with the largest energy dissipation and the largest turbulence intensity are selected as the upstream and downstream sections for calculating the discrete coefficient of water surface lateral gradient.

3. The method for optimizing water flow in a curved channel based on the analysis of rough strip diversion effect evaluation indexes according to claim 1, characterized in that: The selection of the upstream section and the downstream section for calculating the average velocity dispersion coefficient in S2 is determined based on the conservation of water flow momentum and vorticity distribution in the bend, as follows: The velocity of the water flow in the bend is measured using a velocity meter to obtain time-varying velocity data, and the momentum change of the water flow at different positions is calculated using the momentum conservation equation. The vorticity calculation method is also used to determine the distribution of vorticity. Construct a coupled model of water flow momentum and vorticity in a bend to analyze the momentum transfer process in the bend and the influence of vorticity on flow velocity changes; Combined with the analysis results of the coupling model and the vorticity distribution, the two sections with the largest velocity gradient and the largest vorticity value are selected as the upstream and downstream sections for calculating the average velocity dispersion coefficient.

4. The method for optimizing water flow in a curved channel based on the analysis of rough strip diversion effect evaluation indexes according to claim 1, characterized in that: The dynamic grouping method is used in S3 to construct the original data matrix, as follows: According to the actual flow characteristics of the bend water flow, the rough strip arrangement area is divided into a bend entrance section, a bend curvature change section and a bend exit section, and the flow characteristics are determined by analyzing the flow velocity distribution of different sections; According to the relative height differences of rough strips in different sub-areas, the measured water surface lateral gradient dispersion coefficient and average flow velocity dispersion coefficient data are grouped, and for each group, the change rate of the internal data is calculated as a dynamic change feature; According to the dynamic change characteristics, a weight coefficient is assigned to each group, and the weighted group data are arranged and combined to construct the original data matrix.

5. The method for optimizing water flow in a curved channel based on the analysis of rough strip diversion effect evaluation indexes according to claim 1, characterized in that: In S3, the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient are normalized and normalized using an adaptive processing algorithm, which is as follows: The trend decomposition method is used to decompose the coefficient data into trend terms, period terms and random terms, which are used to analyze the time series trend of the measurement data; For trend items, determine the direction and magnitude of positive conversion based on their growth or decline trend; Calculate the period length and fluctuation amplitude of the periodic term and scale the periodic term accordingly; By calculating the probability distribution characteristics of random items, a linear change algorithm is used to adjust their distribution to an approximate normal distribution; According to the overall fluctuation range and distribution pattern of the data, the standardized scaling factor is dynamically determined, and the data is standardized according to the scaling factor.

6. The method for optimizing water flow in a curved channel based on the analysis of rough strip diversion effect evaluation indexes according to claim 1, characterized in that: The modified calculation process for calculating the information carrying capacity of each indicator in S3 is as follows: Based on the measured water depth and flow velocity data, the kinetic energy, potential energy and energy loss of the bend water flow at different positions are calculated. Based on the energy conversion, the influence of rough strips on the water flow energy conversion process is analyzed, and the quantitative relationship between the energy conversion efficiency and the water surface horizontal gradient dispersion coefficient and the average flow velocity dispersion coefficient is determined. Based on the quantitative relationship, the coefficient data is energy corrected. Calculate the momentum components of the water flow in different directions in the bend, analyze the changes in the path and size of the water flow momentum transfer caused by the rough strip, establish a momentum transfer model to obtain the correlation between the momentum transfer characteristic parameters and the coefficients, and perform momentum correction on the coefficients based on the correlation; The coefficient data after energy correction and momentum correction are substituted into the comparative and contradictory formulas for calculating information carrying capacity. At the same time, the Pearson correlation coefficient with energy and momentum weighting is adjusted to calculate the information carrying capacity. Finally, the weights of the two indicators in the comprehensive evaluation are determined based on the calculated information carrying capacity.

7. The method for optimizing water flow in a curved channel based on the analysis of rough strip diversion effect evaluation indexes according to claim 1, characterized in that: When the multi-index comprehensive weighted method is used in S4 to calculate the diversion speed reduction weighted coefficient, a dynamic adjustment coefficient based on water flow energy balance is introduced. The specific steps are as follows: Calculate the kinetic energy and potential energy of the bend water flow at different positions and the amount of energy dissipated by the rough strips, establish the water flow energy balance equation, and analyze the coupling degree between the water surface horizontal gradient discrete coefficient and the average flow velocity discrete coefficient and the water flow energy based on the energy balance relationship to determine the energy influencing factor; A dynamic adjustment coefficient model is constructed based on the energy influencing factor. When the uneven distribution of water flow energy exceeds the preset threshold, the adjustment coefficient of the water surface horizontal gradient discrete coefficient is increased according to the preset proportion, and the dynamic adjustment coefficient of the average flow velocity discrete coefficient is adjusted according to the same proportion. The dynamic adjustment coefficient is combined with the fixed weighting coefficient based on the weight calculation result to obtain the corrected comprehensive weighting coefficient, and the diversion and speed reduction weighting coefficient is calculated.

8. The method for optimizing water flow in a curved channel based on the analysis of rough strip diversion effect evaluation indexes according to claim 1, characterized in that: When the rough stripe arrangement parameters are adjusted and optimized in S4, an optimization method based on a combination of a genetic algorithm and a local search strategy is adopted, which is specifically as follows: The relative height of the rough strips is encoded to form chromosome individuals, and the population is initialized. The weighted coefficient of the diversion speed reduction is used as the fitness function to calculate the fitness value of each individual in the population. The selection, crossover and mutation operations of the genetic algorithm are used to iteratively evolve the population and find the global optimal solution area. In each generation of evolution, the individuals with the highest fitness are retained, the excellent genes between individuals are exchanged through the crossover operation, and new genes are introduced through the mutation operation. When the genetic algorithm iterates to the preset generation, the gradient descent method is used to search the current optimal solution area. In the local search process, the relative height of the rough strips is optimized with the diversion speed reduction weighted coefficient, the water surface horizontal gradient discrete coefficient and the average flow velocity discrete coefficient as the target, until the optimal rough strip relative height combination is found that maximizes the diversion speed reduction weighted coefficient and minimizes the water surface horizontal gradient discrete coefficient and the average flow velocity discrete coefficient.

Images