CN113762610A - Method, system and equipment for predicting maximum development potential of channel bearing capacity - Google Patents
Method, system and equipment for predicting maximum development potential of channel bearing capacity Download PDFInfo
- Publication number
- CN113762610A CN113762610A CN202111000933.7A CN202111000933A CN113762610A CN 113762610 A CN113762610 A CN 113762610A CN 202111000933 A CN202111000933 A CN 202111000933A CN 113762610 A CN113762610 A CN 113762610A
- Authority
- CN
- China
- Prior art keywords
- individual
- individuals
- bearing capacity
- optimal
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000011161 development Methods 0.000 title claims abstract description 67
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 68
- 238000013528 artificial neural network Methods 0.000 claims abstract description 51
- 238000011156 evaluation Methods 0.000 claims abstract description 21
- 230000002068 genetic effect Effects 0.000 claims abstract description 19
- 238000012549 training Methods 0.000 claims abstract description 19
- 230000035772 mutation Effects 0.000 claims description 21
- 238000012937 correction Methods 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 6
- 230000001174 ascending effect Effects 0.000 claims description 4
- 238000012847 principal component analysis method Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 3
- 238000004364 calculation method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 210000000349 chromosome Anatomy 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 4
- VYZAMTAEIAYCRO-UHFFFAOYSA-N Chromium Chemical compound [Cr] VYZAMTAEIAYCRO-UHFFFAOYSA-N 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 101100390771 Danio rerio fitm1l gene Proteins 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 1
- 238000009396 hybridization Methods 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
Abstract
The application relates to a method, equipment and a storage medium for predicting maximum development potential of road bearing capacity, wherein the method comprises the following steps: collecting multiple index evaluation data of different river reach and corresponding development scheme data, and establishing a sample data set; optimizing the weight and the threshold of the BP neural network by adopting a genetic algorithm to obtain an optimal weight and an optimal threshold; constructing a BP neural network, and initializing the BP neural network by using the optimal weight and the optimal threshold; training the initialized BP neural network by using the sample data set to obtain a prediction model; and predicting the maximum development potential of the channel bearing capacity of the river reach to be detected by using the prediction model. The method and the device have the technical effects of high prediction precision and independence on artificial experience prediction.
Description
Technical Field
The present application relates to the field of channel bearing capacity prediction technologies, and in particular, to a method, a system, a device, and a storage medium for predicting a maximum development potential of channel bearing capacity.
Background
In 2016, a concept of 'channel bearing capacity' is firstly put forward domestically, the channel bearing capacity is based on the maximum developable potential of natural endowments, a plurality of factors such as economy, water resource hybridization, flood control, ecology and the like are considered, and the channel developable scale is comprehensively determined, and the maximum developable scale is the maximum developable potential of the channel bearing capacity.
The current research on bearing capacity mainly focuses on the channel passing capacity, and the research on channel depth, channel width and turning radius is less. The maximum exploitable potential of the Xijiang channel under ecological protection is researched by Von Lin, the channel depth is estimated by utilizing a stable channel depth estimation algorithm, the research on the channel depth, the channel width and the turning radius is less, different channel scale schemes are set mainly through human experience, the best scheme is determined through an analytic hierarchy process, and the research on the channel scale is greatly limited. In summary, there is still much research space for the maximum exploitable potential of the bearing capacity of the channel.
Disclosure of Invention
In view of this, the present application provides a method, a system, a device and a storage medium for predicting the maximum development potential of the channel bearing capacity, so as to solve the technical problems that the prediction of the channel bearing capacity is affected by the finite of human experience and the number of schemes, and the prediction precision is low.
In order to solve the above problem, in a first aspect, the present invention provides a method for predicting maximum development potential of channel bearing capacity, including the following steps:
collecting multiple index evaluation data of different river reach and corresponding development scheme data, and establishing a sample data set;
optimizing the weight and the threshold of the BP neural network by adopting a genetic algorithm to obtain an optimal weight and an optimal threshold;
constructing a BP neural network, and initializing the BP neural network by using the optimal weight and the optimal threshold;
training the initialized BP neural network by using the sample data set to obtain a prediction model;
and predicting the maximum development potential of the channel bearing capacity of the river reach to be detected by using the prediction model.
Optionally, collecting multiple index evaluation data of different river reach and corresponding development scheme data, and establishing a sample data set, specifically:
screening indexes influencing the bearing capacity of the channel by adopting a principal component analysis method to obtain various key indexes;
collecting key index evaluation data of different river reach and corresponding development scheme data;
and screening out the collected key index evaluation data and non-stationary data in the corresponding development scheme data, and carrying out normalization processing to obtain a sample data set.
Optionally, a genetic algorithm is adopted to optimize the weight and the threshold of the BP neural network to obtain an optimal weight and an optimal threshold, specifically:
generating an initial weight and an initial threshold by a random number method, and coding the initial weight and the initial threshold to generate an initial population;
optimizing the current population by using a two-edge successive correction method, and selecting an elite individual from the optimized population according to the applicability value to obtain an elite population;
performing cross variation on individuals in the elite population based on a roulette method to obtain a new population, and recording the optimal individual of the iteration;
judging whether the iteration termination condition is met, if so, outputting the optimal individual, and if not, performing the next iteration, wherein the weight and the threshold corresponding to the optimal individual are the optimal weight and the optimal threshold.
Optionally, the current population is optimized by using a two-edge successive correction method, specifically:
selecting individuals needing improvement from the current population;
randomly selecting improvement points, and performing reverse order arrangement on the improvement points in the individual to be improved;
and judging whether all the improvement points are traversed or not, if so, outputting an improved individual, if not, judging whether the individual is improved or not according to the fitness value, if so, outputting the improved individual, and if not, reselecting the improvement points for improvement.
Optionally, selecting an elite individual from the optimized population according to the applicability value to obtain an elite population, specifically:
and taking the average value of the prediction error values of various indexes as a fitness value, sequencing the individuals in an ascending order according to the fitness value, and selecting the individuals in the front in a set proportion as elite individuals to obtain the elite population.
Optionally, performing cross variation on individuals in the elite population based on roulette, including:
performing cross operation on individuals in the elite population based on the improved cross operator;
carrying out cross operation on individuals in the elite population based on an improved cross operator, which specifically comprises the following steps:
selecting two individuals needing cross operation from the elite population, wherein the two individuals are a first individual and a second individual respectively, and calculating the fitness average value of the two individuals to be crossed;
judging whether the fitness average value of two individuals to be crossed is smaller than the fitness average value of the current population, if so, setting the crossing length as a first length, otherwise, setting the crossing length as a second length, wherein the first length is larger than the second length;
randomly generating a starting position of the crossover operation based on the crossover length;
based on the starting position and the crossing length, respectively extracting crossing sequences from two individuals to be crossed;
placing the cross sequence of the second individual at the head end of the first individual, and cutting off the sequence with the same length as the cross length at the tail end of the first individual to obtain the first individual after cross operation;
and placing the cross sequence of the first individual at the head end of the second individual, and cutting off the sequence with the same length as the cross length at the tail end of the second individual to obtain the second individual after the cross operation.
Optionally, performing cross variation on individuals in the elite population based on roulette, including:
carrying out mutation operation on individuals in the elite population based on the improved crossover operator;
carrying out mutation operation on individuals in the elite population based on an improved crossover operator, which specifically comprises the following steps:
selecting individuals needing mutation operation from the elite population;
calculating the fitness value of the individual to be mutated, judging whether the fitness value of the individual to be mutated is larger than the fitness average value of the current population, if so, setting the mutation rate as a first numerical value, otherwise, setting the mutation rate as a second numerical value, wherein the first numerical value is larger than the second numerical value;
randomly determining a variation position;
and carrying out variation operation on the variation position of the individual to be varied according to a set variation formula to obtain the varied individual.
In a second aspect, the present application further provides a system for predicting maximum development potential of channel bearing capacity, the system comprising:
the sample data module is used for collecting index evaluation data of different river reach and corresponding development scheme data and establishing a sample data set;
the genetic algorithm module is used for optimizing the weight and the threshold of the BP neural network by adopting a genetic algorithm to obtain an optimal weight and an optimal threshold;
the neural network module is used for constructing a BP neural network and initializing the BP neural network by using the optimal weight and the optimal threshold; the method is also used for training the initialized BP neural network by using the sample data set to obtain a prediction model
And the prediction module is used for predicting the maximum development potential of the channel bearing capacity of the river reach to be detected by using the prediction model.
In a third aspect, the present application provides a computer device, which adopts the following technical solution:
a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method for predicting the maximum development potential of channel bearing capacity when executing the computer program.
In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solutions:
a computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for predicting the maximum development potential for bearing capacity of a waterway.
The invention has the beneficial effects that: the method adopts a genetic optimization BP neural network algorithm to predict the bearing capacity of the navigation channel, specifically, adopts the genetic algorithm to optimize the initialization weight and the initialization threshold of the BP neural network, accelerates the training speed, trains the initialized BP neural network through the collected sample data set, and can obtain a prediction model for predicting the bearing capacity of the navigation channel. The prediction model is adopted to predict the bearing capacity of the navigation channel, so that the influence of the finite of human experience and the set number of schemes on the prediction result can be reduced, and the prediction precision of the maximum developable potential of the bearing capacity of the navigation channel is improved.
Drawings
Fig. 1 is a flowchart of a method of predicting the maximum development potential of the bearing capacity of the channel according to an embodiment of the present disclosure;
FIG. 2 is a chromosome coding diagram of an embodiment of a method for predicting the maximum development potential of channel bearing capacity provided by the present application;
FIG. 3a is a fitting graph of an embodiment of a method for predicting the maximum development potential of channel bearing capacity provided by the present application;
FIG. 3b is an error diagram of an embodiment of a method for predicting the maximum development potential of channel bearing capacity provided by the present application;
FIG. 3c is an iterative evolutionary diagram of an embodiment of a method for predicting the maximum development potential of channel bearing capacity provided by the present application;
FIG. 4 is a schematic block diagram of an embodiment of a system for predicting maximum development potential of channel bearing capacity provided herein;
FIG. 5 is a schematic block diagram of an embodiment of a computer device provided herein.
Detailed Description
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate preferred embodiments of the application and together with the description, serve to explain the principles of the application and not to limit the scope of the application.
In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
The application provides a method, a system, equipment, computer equipment and a storage medium for predicting the maximum development potential of channel bearing capacity, which are respectively described in detail below.
First, as shown in fig. 1, an embodiment of the present application provides a method for predicting a maximum development potential of a channel bearing capacity, including the following steps:
s1, collecting multiple index evaluation data of different river reach and corresponding development scheme data, and establishing a sample data set;
s2, optimizing the weight and the threshold of the BP neural network by adopting a genetic algorithm to obtain an optimal weight and an optimal threshold;
s3, constructing a BP neural network, and initializing the BP neural network by using the optimal weight and the optimal threshold;
s4, training the initialized BP neural network by using the sample data set to obtain a prediction model;
and S5, predicting the maximum development potential of the channel bearing capacity of the river reach to be detected by using the prediction model.
Specifically, firstly, initializing algorithm parameters, setting the number of nodes of an input layer, a hidden layer and an output layer, and setting parameters such as iteration times, population scale, learning rate and the like; collecting a sample data set for BP neural network training; optimizing the weight and the threshold of the BP neural network by adopting a genetic algorithm to obtain an optimal weight and an optimal threshold; the optimal weight value and the optimal threshold value are used for initializing the BP neural network; and training the initialized BP neural network by using the sample data set to obtain a prediction model for predicting the bearing capacity of the navigation channel.
In the embodiment, the channel bearing capacity is predicted by adopting a genetic optimization BP neural network algorithm, which is referred to as GA-BP algorithm for short. The GA-BP algorithm is generally used for the aspects of risk assessment, risk prediction and the like, is less applied to the aspect of channel bearing capacity prediction, and the influence of the limitation of artificial experience and scheme setting quantity on the prediction result can be effectively reduced and the prediction precision of the maximum developable potential of the channel bearing capacity is improved by applying the GA-BP algorithm to the aspect of the maximum developable potential of the channel bearing capacity.
In one embodiment, multiple index evaluation data of different river reach and corresponding development scheme data are collected, and a sample data set is established, specifically:
screening indexes influencing the bearing capacity of the channel by adopting a principal component analysis method to obtain various key indexes;
collecting key index evaluation data of different river reach and corresponding development scheme data;
and screening out the collected key index evaluation data and non-stationary data in the corresponding development scheme data, and carrying out normalization processing to obtain a sample data set.
And carrying out principal component analysis on the indexes which influence the maximum developable potential of the channel bearing capacity, screening out the key indexes which influence the maximum developable potential of the channel bearing capacity, collecting the key index evaluation data of different river reach and the corresponding development scheme data, and effectively reducing the prediction error caused by high correlation among the indexes. And then, performing stability analysis on the collected original data, removing unstable data and avoiding prediction errors caused by unstable data. And finally, carrying out normalization processing on the data to improve the convergence speed of the algorithm.
In one embodiment, a genetic algorithm is used to optimize the weight and the threshold of the BP neural network to obtain an optimal weight and an optimal threshold, specifically:
generating an initial weight and an initial threshold by a random number method, and coding the initial weight and the initial threshold to generate an initial population;
optimizing the current population by using a two-edge successive correction method, and selecting an elite individual from the optimized population according to the applicability value to obtain an elite population;
performing cross variation on individuals in the elite population based on a roulette method to obtain a new population, and recording the optimal individual of the iteration;
judging whether the iteration termination condition is met, if so, outputting the optimal individual, and if not, performing the next iteration, wherein the weight and the threshold corresponding to the optimal individual are the optimal weight and the optimal threshold.
Generating an initial population representing a combination of a network weight and a threshold value by a random number method, optimizing the initial population by utilizing a two-side successive correction algorithm, taking a prediction error average value of the navigation depth, the navigation width and the turning radius as a fitness value, and screening an elite population based on the fitness value; based on a roulette method, carrying out cross variation on individuals needing cross variation by using an improved cross variation operator, improving the diversity of population individuals, adding a new individual into an elite population, expanding the elite population to an initial population scale, recording the optimal individual of the iteration, namely indicial. chrom (1): chrom _ best (I), judging that the iteration number NC is greater than 1, outputting the optimal individual if the iteration number NC is greater than 1, judging whether an iteration termination condition is met if the iteration number NC is not greater than 1, outputting the optimal individual if the iteration number NC is met, and carrying out the next iteration if the iteration number NC is not greater than 1.
In the embodiment, based on the traditional GA-BP algorithm, two successive correction algorithms are embedded, so that the algorithm search area is enlarged, and the algorithm convergence speed is improved.
Specifically, the weight and the threshold of the BP neural network are real numbers, so the algorithm adopts real number coding, adopts a single chromosome coding mode, codes the weight and the threshold together, splits the optimal solution, namely the chromosome of the optimal individual into the weight and the threshold corresponding to each edge and node of the network, and then predicts by using the BP network. The coding is illustrated by a network of 3 inputs, 3 outputs and 1 hidden layer node, and the coding diagram is shown in fig. 2. And (3) decoding: and segmenting and intercepting the chromosome according to the corresponding relation between the network edge and the node, substituting the optimized data into a BP network as a network weight and a threshold, performing fitting calculation on the data, and outputting a prediction result and a fitting error.
In an embodiment, a two-edge successive correction method is used to optimize a current population, specifically:
selecting individuals needing improvement from the current population;
randomly selecting improvement points, and performing reverse order arrangement on the improvement points in the individual to be improved;
and judging whether all the improvement points are traversed or not, if so, outputting an improved individual, if not, judging whether the individual is improved or not according to the fitness value, if so, outputting the improved individual, and if not, reselecting the improvement points for improvement.
Local optimization is carried out on the elite individual by utilizing a two-edge successive correction method, which comprises the following steps: inputting an individual in need of improvement; randomly selecting an improved point to make individual1(i + 1: j) equal to individual1 (j: 1: i +1), finishing reverse order arrangement, and recording a new path as individual 1; judging whether the improvement points are traversed or not, and outputting an improved individual1 if the traversal is finished; and judging whether the individual fitness value is improved or not if the individual fitness value is not completely traversed, if so, enabling i to be i +1, outputting an improved individual, and if not, reselecting an improved point to continue improvement.
In one embodiment, selecting elite individuals from the optimized population according to the applicability value to obtain an elite population, specifically:
and taking the average value of the prediction error values of various indexes as a fitness value, sequencing the individuals in an ascending order according to the fitness value, and selecting the individuals in the front in a set proportion as elite individuals to obtain the elite population.
In this embodiment, the average values of the prediction errors of the navigation depth, the navigation width and the turning radius are taken as fitness values, the fitness values are arranged in an ascending order, and the individuals in the top 30% are selected as elite populations.
In one embodiment, the cross-variation of individuals within the elite population based on roulette comprises:
performing cross operation on individuals in the elite population based on the improved cross operator;
carrying out cross operation on individuals in the elite population based on an improved cross operator, which specifically comprises the following steps:
selecting two individuals needing cross operation from the elite population, wherein the two individuals are a first individual and a second individual respectively, and calculating the fitness average value of the two individuals to be crossed;
judging whether the fitness average value of two individuals to be crossed is smaller than the fitness average value of the current population, if so, setting the crossing length as a first length, otherwise, setting the crossing length as a second length, wherein the first length is larger than the second length;
randomly generating a starting position of the crossover operation based on the crossover length;
based on the starting position and the crossing length, respectively extracting crossing sequences from two individuals to be crossed;
placing the cross sequence of the second individual at the head end of the first individual, and cutting off the sequence with the same length as the cross length at the tail end of the first individual to obtain the first individual after cross operation;
and placing the cross sequence of the first individual at the head end of the second individual, and cutting off the sequence with the same length as the cross length at the tail end of the second individual to obtain the second individual after the cross operation.
Specifically, two individuals, namely indivisual 1 and indivisual 2, which need to be crossed are input, the average value of the fitness of indivisual 1 and indivisual 2 is calculated and is marked as fit12, and the average value is compared with the current population average fitness fitavg; if fit12< fitavg, the crossing length is 5, otherwise, the crossing length is 2, the crossing length is marked as nn, step4 is switched; randomly generating a starting position by a randderm function, determining a starting position ps of the intersection, wherein ps is randderm (n-nn); placing the cross-over sequence of the Individual2 at the left end of the Individual1, cutting off the same length sequence at the tail end of the recombinant Individual, namely Individual1 ═ Individual2 (ps: ps + nn-1), Individual1 (1: n-nn) ], treating the same Individual2 in a way of Individual2 ═ Individual1 (ps: ps + nn-1), Individual2 (1: n-nn) ], and outputting the new Individual.
The embodiment improves the crossover operator, adjusts the crossover length based on the fitness value, and simultaneously, performs region crossover, improves the stability of the population, and is beneficial to the algorithm to realize global optimization.
In one embodiment, the cross-variation of individuals within the elite population based on roulette comprises:
carrying out mutation operation on individuals in the elite population based on the improved crossover operator;
carrying out mutation operation on individuals in the elite population based on an improved crossover operator, which specifically comprises the following steps:
selecting individuals needing mutation operation from the elite population;
calculating the fitness value of the individual to be mutated, judging whether the fitness value of the individual to be mutated is larger than the fitness average value of the current population, if so, setting the mutation rate as a first numerical value, otherwise, setting the mutation rate as a second numerical value, wherein the first numerical value is larger than the second numerical value;
randomly determining a variation position;
and carrying out variation operation on the variation position of the individual to be varied according to a set variation formula to obtain the varied individual.
Specifically, an individual requiring variation is input into individual1, and the fitness value fit1 and the population fitness average value fitavg of the individual1 are calculated; if fit1> fitavg, adjusting the variation rate pm to be 0.15, otherwise, adjusting the variation rate pm to be 0.08; determining a variation position by using ps (randderm) (n) and n as the length of the chromosome; let individual1(ps) — 6 × rand +3, rand denote a random number; and outputting the new individual.
The embodiment improves the mutation operator, adjusts the mutation rate based on the fitness value, is favorable for improving the stability of the population, and is favorable for realizing the global optimum of the algorithm.
In order to verify the effect of the improved GA-BP algorithm of the invention, the following steps are respectively carried out on the same data by utilizing the traditional BP algorithm, the GA-BP algorithm and the optimized GA-BP algorithm of the invention, and the calculation results are compared.
Index evaluation data of different river reach of the upstream, middle and downstream of the project and corresponding development scheme results are respectively used as input and output, a sample data set is established, and the sample data set is divided into a training data group and a test data group. And establishing a multi-input multi-output BP network. And training the improved GA-BP algorithm by utilizing a training data set, and finally testing the effect of the training network by using test data, wherein the trained network can be used for predicting the maximum developable potential of the channel bearing capacity of different river reach. The data of each index part of different river reach are shown in the table 1.
TABLE 1 evaluation data and results Table
In order to improve the convergence rate of the algorithm, the raw data is normalized by using a mapmaxmin function, and the result of the normalization process is shown in table 2.
Table 2, data normalization processing result table
The algorithm operating environment is CPU2.20GHz, the memory is 4.00GB, the operating system is 64 bits Windows10, and the programming language adopts MATLABR2016 a.
Algorithm parameters are set, the iteration number G is 500, the population size NP is 100, the training number net.trainParam.epochs is 20, the learning rate net.trainParam.lr is 0.1, the training target minimum error net.trainParam.goal is 0.00001, and the reality frequency net.trainParam.show is 100.
Based on the idea of control variables, the same data are respectively calculated by using the traditional GA-BP algorithm and the optimized GA-BP algorithm, wherein the parameters of the BP are selected to be consistent with the parameters in the improved GA-BP, the traditional GA-BP algorithm selection module is carried out by adopting a roulette method, the cross mutation part adopts single-point co-location cross and single-point mutation, and the cross rate and the mutation rate are respectively 0.8 and 0.008. The results of the different algorithm calculations are compared in table 3, where D denotes the channel depth, B denotes the channel width, and R denotes the turning radius.
TABLE 3 comparison of calculation results of different algorithms
The fitting graph, the error graph and the iterative evolution graph of the improved GA-BP algorithm to the data are respectively shown in fig. 3a, fig. 3b and fig. 3 c.
And carrying out variance analysis on the data fitting results of different algorithms, and comparing the stability of the data fitting of the different algorithms. The variance calculation formula is as follows:
wherein S is2Representing variance, n representing number of samples, SiRepresents the algorithm fit value, TiRepresenting the true value.
The variance of the fit of different algorithms to the depth of flight, the width of flight and the turning radius is shown in table 4.
TABLE 4 table of variance results of different algorithms fitting to test data
As can be seen from Table 3, the fitting accuracy of the improved GA-BP algorithm to data is higher than that of the conventional GA-BP algorithm and BP algorithm, and as can be seen from Table 4, the improved GA-BP algorithm has smaller variance and higher stability to data fitting compared with the conventional GA-BP algorithm and BP algorithm.
In conclusion, the improved GA-BP algorithm has high prediction precision on the maximum developable potential of the channel bearing capacity and good stability. Compared with the traditional calculation method for obtaining the maximum developable potential of the bearing capacity of the navigation channel through the analytic hierarchy process comparison scheme based on the proposed scheme, the method saves the labor cost, improves the prediction precision, and can more effectively predict the maximum developable potential of the bearing capacity of the navigation channel. Compared with the traditional GA-BP algorithm and BP algorithm, the improved GA-BP algorithm has smaller prediction error and higher stability. Therefore, the proposed improved GA-BP algorithm can be used for predicting the maximum developable potential of the bearing capacity of the navigation channel.
The invention embeds two-side successive correction algorithm, improved crossover operator and improved mutation operator on the basis of the traditional GA-BP algorithm, and provides an improved GA-BP algorithm which can predict the maximum developable potential of the bearing capacity of the channels of different river reach. And (3) screening key indexes influencing the maximum developable potential of the bearing capacity of the navigation channel through principal component analysis, removing unstable components in the original data, and using the screened data for improving the training and testing of the GA-BP algorithm. The improved GA-BP algorithm has small prediction error and high stability. The method is suitable for predicting the maximum developable potential of the bearing capacity of the navigation channel.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
The embodiment also provides a prediction system of the maximum development potential of the bearing capacity of the channel, and the prediction system of the maximum development potential of the bearing capacity of the channel corresponds to the prediction method of the maximum development potential of the bearing capacity of the channel in the embodiment one to one. As shown in fig. 4, the system for predicting the maximum development potential of channel bearing capacity includes:
the sample data module 401 is configured to collect index evaluation data of different river reach and corresponding development scheme data, and establish a sample data set;
a genetic algorithm module 402, configured to optimize the weight and the threshold of the BP neural network by using a genetic algorithm to obtain an optimal weight and an optimal threshold;
the neural network module 403 is configured to construct a BP neural network, and initialize the BP neural network by using the optimal weight and the optimal threshold; the method is also used for training the initialized BP neural network by using the sample data set to obtain a prediction model
And the prediction module 404 is configured to predict the maximum development potential of the channel bearing capacity of the river reach to be detected by using the prediction model.
For specific limitations of the prediction system regarding the maximum development potential of the bearing capacity of the channel, see above for
The definition of the method is not described herein. All or part of each module in the channel bearing capacity maximum development potential prediction system can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
As shown in fig. 5, based on the method for predicting the maximum development potential of the bearing capacity of the channel, the present application also provides a computer device, which may be a mobile terminal, a desktop computer, a notebook computer, a palm computer, a server, or other computing devices. The computer device comprises a processor 10, a memory 20 and a display 30. FIG. 5 shows only some of the components of a computer device, but it is to be understood that not all of the shown components are required to be implemented, and that more or fewer components may be implemented instead.
The storage 20 may in some embodiments be an internal storage unit of the computer device, such as a hard disk or a memory of the computer device. The memory 20 may also be an external storage device of the computer device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), etc. provided on the computer device. Further, the memory 20 may also include both an internal storage unit and an external storage device of the computer device. The memory 20 is used for storing application software installed in the computer device and various data, such as program codes installed in the computer device. The memory 20 may also be used to temporarily store data that has been output or is to be output. In an embodiment, the memory 20 stores a channel bearing capacity maximum development potential prediction program 40, and the channel bearing capacity maximum development potential prediction program 40 can be executed by the processor 10, so as to implement the channel bearing capacity maximum development potential prediction method according to the embodiments of the present application.
The processor 10 may be, in some embodiments, a Central Processing Unit (CPU), microprocessor or other data Processing chip for executing program codes stored in the memory 20 or Processing data, such as performing a method for predicting maximum development potential of channel bearing capacity.
The display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch panel, or the like in some embodiments. The display 30 is used for displaying information at the computer device and for displaying a visual user interface. The components 10-30 of the computer device communicate with each other via a system bus.
In one embodiment, the following steps are implemented when processor 10 executes a channel bearing capacity maximum development potential prediction program 40 in memory 20:
collecting multiple index evaluation data of different river reach and corresponding development scheme data, and establishing a sample data set;
optimizing the weight and the threshold of the BP neural network by adopting a genetic algorithm to obtain an optimal weight and an optimal threshold;
constructing a BP neural network, and initializing the BP neural network by using the optimal weight and the optimal threshold;
training the initialized BP neural network by using the sample data set to obtain a prediction model;
and predicting the maximum development potential of the channel bearing capacity of the river reach to be detected by using the prediction model.
The present embodiment also provides a computer-readable storage medium, on which a program for predicting a maximum development potential of a bearing capacity of a waterway is stored, the program implementing the following steps when executed by a processor:
collecting multiple index evaluation data of different river reach and corresponding development scheme data, and establishing a sample data set;
optimizing the weight and the threshold of the BP neural network by adopting a genetic algorithm to obtain an optimal weight and an optimal threshold;
constructing a BP neural network, and initializing the BP neural network by using the optimal weight and the optimal threshold;
training the initialized BP neural network by using the sample data set to obtain a prediction model;
and predicting the maximum development potential of the channel bearing capacity of the river reach to be detected by using the prediction model.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above.
Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The above description is only for the preferred embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application.
Claims (10)
1. A prediction method for the maximum development potential of channel bearing capacity is characterized by comprising the following steps:
collecting multiple index evaluation data of different river reach and corresponding development scheme data, and establishing a sample data set;
optimizing the weight and the threshold of the BP neural network by adopting a genetic algorithm to obtain an optimal weight and an optimal threshold;
constructing a BP neural network, and initializing the BP neural network by using the optimal weight and the optimal threshold;
training the initialized BP neural network by using the sample data set to obtain a prediction model;
and predicting the maximum development potential of the channel bearing capacity of the river reach to be detected by using the prediction model.
2. The method for predicting the maximum development potential of the bearing capacity of the channel according to claim 1, wherein a sample data set is established by collecting multiple index evaluation data of different river reach and corresponding development scheme data, and specifically comprises the following steps:
screening indexes influencing the bearing capacity of the channel by adopting a principal component analysis method to obtain various key indexes;
collecting key index evaluation data of different river reach and corresponding development scheme data;
and screening out the collected key index evaluation data and non-stationary data in the corresponding development scheme data, and carrying out normalization processing to obtain a sample data set.
3. The method for predicting the maximum development potential of the bearing capacity of the channel according to claim 1, wherein a genetic algorithm is adopted to optimize the weight and the threshold of the BP neural network to obtain an optimal weight and an optimal threshold, and the method specifically comprises the following steps:
generating an initial weight and an initial threshold by a random number method, and coding the initial weight and the initial threshold to generate an initial population;
optimizing the current population by using a two-edge successive correction method, and selecting an elite individual from the optimized population according to the applicability value to obtain an elite population;
performing cross variation on individuals in the elite population based on a roulette method to obtain a new population, and recording the optimal individual of the iteration;
judging whether the iteration termination condition is met, if so, outputting the optimal individual, and if not, performing the next iteration, wherein the weight and the threshold corresponding to the optimal individual are the optimal weight and the optimal threshold.
4. The method for predicting the maximum development potential of the bearing capacity of the channel according to claim 3, wherein the current population is optimized by using a two-edge successive correction method, and specifically comprises the following steps:
selecting individuals needing improvement from the current population;
randomly selecting improvement points, and performing reverse order arrangement on the improvement points in the individual to be improved;
and judging whether all the improvement points are traversed or not, if so, outputting an improved individual, if not, judging whether the individual is improved or not according to the fitness value, if so, outputting the improved individual, and if not, reselecting the improvement points for improvement.
5. The method for predicting the maximum development potential of the bearing capacity of the channel according to claim 3, wherein the elite individual is selected from the optimized population according to the applicability value to obtain an elite population, and the method specifically comprises the following steps:
and taking the average value of the prediction error values of various indexes as a fitness value, sequencing the individuals in an ascending order according to the fitness value, and selecting the individuals in the front in a set proportion as elite individuals to obtain the elite population.
6. The method of claim 3, wherein the cross-variation of individuals in the elite population based on roulette comprises:
performing cross operation on individuals in the elite population based on the improved cross operator;
carrying out cross operation on individuals in the elite population based on an improved cross operator, which specifically comprises the following steps:
selecting two individuals needing cross operation from the elite population, wherein the two individuals are a first individual and a second individual respectively, and calculating the fitness average value of the two individuals to be crossed;
judging whether the fitness average value of two individuals to be crossed is smaller than the fitness average value of the current population, if so, setting the crossing length as a first length, otherwise, setting the crossing length as a second length, wherein the first length is larger than the second length;
randomly generating a starting position of the crossover operation based on the crossover length;
based on the starting position and the crossing length, respectively extracting crossing sequences from two individuals to be crossed;
placing the cross sequence of the second individual at the head end of the first individual, and cutting off the sequence with the same length as the cross length at the tail end of the first individual to obtain the first individual after cross operation;
and placing the cross sequence of the first individual at the head end of the second individual, and cutting off the sequence with the same length as the cross length at the tail end of the second individual to obtain the second individual after the cross operation.
7. The method of claim 3, wherein the cross-variation of individuals in the elite population based on roulette comprises:
carrying out mutation operation on individuals in the elite population based on the improved crossover operator;
carrying out mutation operation on individuals in the elite population based on an improved crossover operator, which specifically comprises the following steps:
selecting individuals needing mutation operation from the elite population;
calculating the fitness value of the individual to be mutated, judging whether the fitness value of the individual to be mutated is larger than the fitness average value of the current population, if so, setting the mutation rate as a first numerical value, otherwise, setting the mutation rate as a second numerical value, wherein the first numerical value is larger than the second numerical value;
randomly determining a variation position;
and carrying out variation operation on the variation position of the individual to be varied according to a set variation formula to obtain the varied individual.
8. A system for predicting maximum development potential of bearing capacity of a navigation channel, the system comprising:
the sample data module is used for collecting index evaluation data of different river reach and corresponding development scheme data and establishing a sample data set;
the genetic algorithm module is used for optimizing the weight and the threshold of the BP neural network by adopting a genetic algorithm to obtain an optimal weight and an optimal threshold;
the neural network module is used for constructing a BP neural network and initializing the BP neural network by using the optimal weight and the optimal threshold; the method is also used for training the initialized BP neural network by using the sample data set to obtain a prediction model
And the prediction module is used for predicting the maximum development potential of the channel bearing capacity of the river reach to be detected by using the prediction model.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method for predicting the maximum development potential for bearing capacity of a waterway according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when being executed by a processor, carries out the steps of the method for predicting the maximum development potential for bearing capacity of a waterway according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111000933.7A CN113762610A (en) | 2021-08-30 | 2021-08-30 | Method, system and equipment for predicting maximum development potential of channel bearing capacity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111000933.7A CN113762610A (en) | 2021-08-30 | 2021-08-30 | Method, system and equipment for predicting maximum development potential of channel bearing capacity |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113762610A true CN113762610A (en) | 2021-12-07 |
Family
ID=78791793
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111000933.7A Pending CN113762610A (en) | 2021-08-30 | 2021-08-30 | Method, system and equipment for predicting maximum development potential of channel bearing capacity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113762610A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103426027A (en) * | 2013-07-24 | 2013-12-04 | 浙江大学 | Intelligent normal pool level optimal selection method based on genetic neural network models |
| CN103971162A (en) * | 2014-04-04 | 2014-08-06 | 华南理工大学 | Method for improving BP (back propagation) neutral network and based on genetic algorithm |
| CN108009639A (en) * | 2017-12-13 | 2018-05-08 | 重庆邮电大学 | A kind of city ecology construction evaluation method based on GA-BP neural network algorithms |
| CN110443418A (en) * | 2019-07-31 | 2019-11-12 | 西安科技大学 | Urban water consumption prediction technique based on GA-BP neural network |
| CN111160520A (en) * | 2019-12-06 | 2020-05-15 | 南京理工大学 | BP neural network wind speed prediction method based on genetic algorithm optimization |
| CN112100927A (en) * | 2020-09-24 | 2020-12-18 | 湖南工业大学 | Prediction method for slope deformation and soft soil foundation settlement based on GA-BP neural network |
-
2021
- 2021-08-30 CN CN202111000933.7A patent/CN113762610A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103426027A (en) * | 2013-07-24 | 2013-12-04 | 浙江大学 | Intelligent normal pool level optimal selection method based on genetic neural network models |
| CN103971162A (en) * | 2014-04-04 | 2014-08-06 | 华南理工大学 | Method for improving BP (back propagation) neutral network and based on genetic algorithm |
| CN108009639A (en) * | 2017-12-13 | 2018-05-08 | 重庆邮电大学 | A kind of city ecology construction evaluation method based on GA-BP neural network algorithms |
| CN110443418A (en) * | 2019-07-31 | 2019-11-12 | 西安科技大学 | Urban water consumption prediction technique based on GA-BP neural network |
| CN111160520A (en) * | 2019-12-06 | 2020-05-15 | 南京理工大学 | BP neural network wind speed prediction method based on genetic algorithm optimization |
| CN112100927A (en) * | 2020-09-24 | 2020-12-18 | 湖南工业大学 | Prediction method for slope deformation and soft soil foundation settlement based on GA-BP neural network |
Non-Patent Citations (3)
| Title |
|---|
| 杨客: "遗传算法优化的 BP 神经网络在连云港港口吞吐量预测中的应用研究", 《中国优秀硕士学位论文全文数据库（工程科技Ⅱ辑）》, vol. 1, no. 07, pages 234 - 235 * |
| 殷勇: "《交通运输设备》", vol. 1, 西南交通大学出版社, pages: 201 * |
| 袁豪: "旅行商问题的研究与应用", 《中国优秀硕士学位论文全文数据库（经济与管理科学辑）》, no. 02, pages 33 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shrikumar et al. | Technical note on transcription factor motif discovery from importance scores (TF-MoDISco) version 0.5. 6.5 | |
| CN112735535B (en) | Prediction model training method, prediction model training device, data prediction method, data prediction device and storage medium | |
| CN112035549B (en) | Data mining method, device, computer equipment and storage medium | |
| CN111258767A (en) | Intelligent cloud computing resource allocation method and device for complex system simulation application | |
| Stahl et al. | EPSILON-CP: using deep learning to combine information from multiple sources for protein contact prediction | |
| CA3112296A1 (en) | Global ancestry determination system | |
| CN112749495A (en) | Multipoint-point-adding-based proxy model optimization method and device and computer equipment | |
| Rodríguez-Fdez et al. | S-FRULER: Scalable fuzzy rule learning through evolution for regression | |
| CN113762350A (en) | Abnormal data detection method and device, computer equipment and storage medium | |
| CN111209930B (en) | Method and device for generating trust policy and electronic equipment | |
| JP2020529060A (en) | Prediction of molecular properties of molecular variants using residue-specific molecular structural features | |
| Ali et al. | Improved differential evolution algorithm with decentralisation of population | |
| CN109767034B (en) | Relay protection constant value optimization method and device, computer equipment and storage medium | |
| Nygård et al. | Partial least squares Cox regression for genome-wide data | |
| CN106294343A (en) | Data clustering method, model fusion method and device | |
| CN107516020B (en) | Method, device, equipment and storage medium for determining importance of sequence sites | |
| Tembine | Mean field stochastic games: Convergence, Q/H-learning and optimality | |
| CN113762610A (en) | Method, system and equipment for predicting maximum development potential of channel bearing capacity | |
| US20230410936A1 (en) | Network approach to navigating the human genome | |
| Wang et al. | Explore the hidden treasure in protein–protein interaction networks—An iterative model for predicting protein functions | |
| CN114446393A (en) | Method, electronic device and computer storage medium for predicting liver cancer feature type | |
| CN113205185A (en) | Network model optimization method and device, computer equipment and storage medium | |
| Bangar et al. | Investigation and performance improvement of web cache recommender system | |
| Wu et al. | Predicting the robustness of real-world complex networks | |
| CN110738318A (en) | Method, system and device for evaluating network structure running time and generating evaluation model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |