CN107085942A - A traffic flow prediction method, device and system based on wolf pack algorithm - Google Patents

Abstract

The embodiment of the invention discloses a kind of traffic flow forecasting method based on wolf pack algorithm, apparatus and system, including obtain traffic flow data；Traffic flow data progress is handled using the wavelet neural network forecasting traffic flow model pre-established and obtains forecasting traffic flow result；Wherein, wavelet neural network forecasting traffic flow model is formed based on wolf pack Algorithm for Training, and its training process is to calculate initialization wavelet neural network parameter according to historical data and wolf pack algorithm；Initialization wavelet neural network parameter is trained using wavelet neural network and historical data and obtains wavelet neural network forecasting traffic flow model.It can be seen that, the embodiment of the present invention improves predetermined speed and precision of prediction to a certain extent when the wavelet neural network forecasting traffic flow model gone out using the initialization wavelet neural network parameter training obtained based on wolf pack algorithm is predicted to traffic flow.

Description

A traffic flow prediction method, device and system based on wolf pack algorithm

technical field

The embodiments of the present invention relate to the technical field of road traffic, in particular to a traffic flow prediction method, device and system based on wolf pack algorithm.

Background technique

When predicting road traffic flow, it is usually affected by factors such as road conditions, time points, weather changes, etc., resulting in high uncertainty in road traffic flow data, and the law is not obvious. In the prior art, the traditional wavelet neural network method is used to train the network parameters of the wavelet neural network when the traffic flow of the road is predicted. The same gradient descent method for the network, and the gradient descent method is unidirectional, and the relevant network parameters are randomly generated, so that the network parameters are extremely easy to fall into the local minimum during the optimization process, so that the traffic flow prediction speed and prediction accuracy reduce.

Therefore, how to provide a wolf pack algorithm-based traffic flow prediction method, device and system that solves the above technical problems has become a problem that those skilled in the art need to solve.

Contents of the invention

The purpose of the embodiment of the present invention is to provide a traffic flow prediction method, device and system based on wolf pack algorithm, which improves the prediction speed and prediction accuracy to a certain extent during the use process.

In order to solve the above technical problems, an embodiment of the present invention provides a traffic flow prediction method based on wolf pack algorithm, the method comprising:

Obtain traffic flow data;

Using the pre-established wavelet neural network traffic flow prediction model to process the traffic flow data to obtain the traffic flow prediction result; wherein, the wavelet neural network traffic flow prediction model is trained based on the wolf pack algorithm, and the training process is as follows: :

Calculate the initial wavelet neural network parameters based on historical data and wolf pack algorithm;

The wavelet neural network traffic flow prediction model is obtained by using the wavelet neural network and the historical data to train the initialized wavelet neural network parameters.

Optionally, the process of calculating and initializing wavelet neural network parameters based on historical data and wolf pack algorithm is specifically:

encoding each network parameter into each individual wolf according to the historical data, and the position of each individual wolf corresponds to each network parameter;

Finding the position of the optimized head wolf from each of the individual wolves according to preset control parameters and corresponding strategies;

The position is decoded to obtain network parameters corresponding to the position, and the network parameters are used as initial wavelet neural network parameters.

Optionally, the preset control parameters include the total number of individual wolves, the maximum number of iterations, the maximum number of walks, the preset number and the preset distance;

The process of finding the position of the optimized alpha wolf from each of the individual wolves according to the preset control parameters and corresponding strategies is specifically as follows:

S2121: Select a preset number of wolves other than the head wolf as wolf detectors, execute the walking behavior improved by the optimization strategy of the law of universal gravitation, and use the position corresponding to the wolf with the highest prey odor concentration in the current pack of wolves as the wolf detector wander. walk direction;

S2122: Update the position of the wolf detection, and until the prey odor concentration of the wolf detection is greater than the prey odor concentration of the head wolf or the current number of wandering reaches the maximum number of travels, replace the corresponding wolf detection position with that of the head wolf position, the detective wolf becomes the new head wolf;

S2123: Summon ferocious wolves through the new alpha wolf, and process the acquired wolf variables by using chaotic initialization to obtain the position of the initialized ferocious wolf;

S2124: Generate the position of the new wolf individual according to the position of the initialized wolf, calculate the prey odor concentration of the new wolf individual, and when the prey odor concentration is greater than the prey odor concentration of the new head wolf, calculate the The position of the new wolf individual replaces the position of the head wolf; until the distance between the new wolf individual and the head wolf is less than the preset distance;

S2125: Execute the siege behavior optimized by the inertia weight adaptive strategy, and update the position of the head wolf;

S2126: Update the position of the wolf pack according to the "winner is king" rule, and then update the wolf pack according to the "survival of the strong" mechanism and the principle of "survival of the strong, the jungle of the jungle";

S2127: Determine whether the preset accuracy is reached between the alpha wolf and the prey or the current number of iterations is greater than the maximum number of iterations, if so, use the alpha wolf as the optimized alpha wolf, and output the position of the optimized alpha wolf ; Otherwise, return to S2121.

Optionally, the process of updating the location of the wolf detection is specifically:

The position of the wolf detection is updated according to the first calculation relation; the first calculation relation is:

Wherein, the x′ _i represents the updated position of the i-th wolf detector; the x _i represents the current position of the i-th wolf detector, the x _k represents the position of the k-th wolf detector, and the rand(0 , 1) represents a uniform distribution function from 0 to 1, the x _best represents the position where the current prey odor concentration is the largest, and the Y _ik represents the prey odor concentration function of the i-th wolf relative to the k-th wolf;

The Y _ik is obtained according to the second calculation relational expression, and the second calculation relational expression is: Wherein, the G represents the gravitational constant; the Y( _xi ) represents the prey odor concentration of the i-th wolf detection, and the Y(x _k ) represents the prey odor concentration of the k-th wolf detection.

Optionally, the process of generating the position of the new wolf individual according to the position of the initialized wolf is specifically as follows:

Obtain the position of the new individual of the wolf according to the position of the initialization wolf and the third calculation relation, and the third calculation relation is:

x' _n =Y _i +R _i (2x _n,k -1), wherein, the x _n 'is the position of the new wolf individual, and the x _n,k represents the position of the new wolf individual at the kth iteration Position; said Y _i is the odor concentration of the prey; said R _i is the radius of the attack area;

The x _n,k is obtained according to the fourth calculation relational expression, and the fourth calculation relational expression is x _n,k+1 =μx _n,k (1-x _n,k ), wherein, n∈[1,N ], the N represents the total number of individual wolves, the μ represents the control parameters of the chaotic state, and the k represents the number of iterations.

Optionally, the process of executing the siege behavior optimized by the inertia weight adaptive strategy and updating the position of the head wolf is as follows:

Execute the siege behavior optimized by the inertia weight adaptive strategy, and update the position of the head wolf according to the fifth calculation relation, the fifth calculation relation is Wherein, said M ^k represents the position of the prey when the number of iterations is k, said step _c represents the attack step size of the wolf, and said γ represents the inertia weight; said γ is obtained according to the sixth calculation relational formula, and the sixth The calculation relation is:

Wherein, the γ _max represents the maximum inertial weight; the γ _min represents the minimum inertial weight, the Z _max represents the maximum number of iterations, and the z represents the current number of iterations.

In order to solve the above technical problems, an embodiment of the present invention provides a traffic flow prediction device based on wolf pack algorithm, the device includes:

An acquisition module, configured to acquire traffic flow data;

A processing module, configured to process the traffic flow data using a pre-established wavelet neural network traffic flow prediction model to obtain a traffic flow prediction result; wherein the wavelet neural network traffic flow prediction model includes:

Calculation module, used to calculate the initial wavelet neural network parameters according to historical data and wolf pack algorithm;

The training module is configured to use the wavelet neural network and the historical data to train the parameters of the initialized wavelet neural network to obtain the traffic flow prediction model of the wavelet neural network.

Optionally, the calculation module includes:

An individual wolf encoding unit, configured to encode each network parameter into each individual wolf according to historical data, and the position of each individual wolf is in one-to-one correspondence with each of the network parameters;

The head wolf finding unit is used to find the optimized position of the head wolf from each of the individual wolves according to preset control parameters and corresponding strategies;

A decoding unit, configured to decode the position to obtain a network parameter corresponding to the position, and use the network parameter as an initialization wavelet neural network parameter.

In order to solve the above technical problems, an embodiment of the present invention provides a traffic flow prediction system based on the wolf pack algorithm, including the traffic flow prediction device based on the wolf pack algorithm as described above.

The embodiment of the present invention provides a traffic flow prediction method, device and system based on wolf pack algorithm, including: obtaining traffic flow data; using a pre-established wavelet neural network traffic flow prediction model to process the traffic flow data to obtain traffic flow prediction Results; Among them, the wavelet neural network traffic flow prediction model is trained based on the wolf pack algorithm, and the training process is to calculate the initial wavelet neural network parameters based on historical data and wolf pack algorithm; use wavelet neural network and historical data to initialize wavelet The neural network parameters are trained to obtain the wavelet neural network traffic flow prediction model.

It can be seen that the initial wavelet neural network parameters of the wavelet neural network traffic flow prediction model adopted in the embodiments of the present invention when predicting traffic flow are calculated based on historical data and the wolf pack algorithm. Because of its wide range, it can converge to the global optimal solution to a large extent. Therefore, when the wavelet neural network traffic flow prediction model trained by the initial wavelet neural network parameters based on the wolf pack algorithm is used to predict the traffic flow, To a certain extent, the prediction speed and prediction accuracy are improved.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the prior art and the accompanying drawings that need to be used in the embodiments. Obviously, the accompanying drawings in the following description are only some of the present invention. Embodiments, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without any creative effort.

Fig. 1 is a schematic flow chart of a traffic flow prediction method based on wolf pack algorithm provided by an embodiment of the present invention;

Fig. 2 adopts the expressway traffic flow prediction simulation schematic diagram of the traffic flow prediction method based on wolf pack algorithm provided by the embodiment of the present invention;

Fig. 3 is the expressway traffic flow prediction emulation schematic diagram that adopts the traffic flow prediction method of the wavelet neural network in the prior art;

FIG. 4 is a schematic structural diagram of a traffic flow prediction device based on wolf pack algorithm provided by an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a wavelet neural network traffic flow prediction model provided by an embodiment of the present invention.

detailed description

The embodiment of the present invention provides a traffic flow prediction method, device and system based on wolf pack algorithm, which improves the prediction speed and prediction accuracy to a certain extent during the use process.

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Please refer to FIG. 1 . FIG. 1 is a schematic flowchart of a traffic flow prediction method based on wolf pack algorithm provided by an embodiment of the present invention. The method includes:

S11: Obtain traffic flow data;

S12: Use the pre-established wavelet neural network traffic flow prediction model to process the traffic flow data to obtain the traffic flow prediction result; wherein, the wavelet neural network traffic flow prediction model is trained based on the wolf pack algorithm, and the training process is as follows:

S21: Calculate the initial wavelet neural network parameters based on historical data and wolf pack algorithm;

S22: Using the wavelet neural network and historical data to train the initial wavelet neural network parameters to obtain a wavelet neural network traffic flow prediction model.

It should be noted that the traffic flow data of roads (such as expressways) are obtained, and the traffic flow prediction is carried out through the pre-established wavelet neural network traffic flow prediction model based on these traffic flow data. The initial wavelet neural network parameters (for example, wavelet neural network connection weights and threshold parameters, etc.) used to train the wavelet neural network traffic flow prediction model in the embodiment of the present invention are calculated by the wolf pack algorithm, for example, can be obtained from The traffic data control center obtains historical data (that is, historical traffic flow data), and uses the wolf pack algorithm for calculation and processing, thereby obtaining the initial wavelet neural network parameters, and then uses the historical data and initial wavelet neural network parameters to be input into the wavelet neural network. The wavelet neural network traffic flow prediction model is obtained through training.

Specifically, in practical applications, for example, for the traffic flow prediction of a certain expressway, the traffic flow data can be obtained from the database of the traffic data control center corresponding to the expressway, and the selected forecast section in May 2016 can be A total of 2976 traffic flow data in 31 days are used as experimental data. In order to make the prediction results more accurate, the acquired original traffic flow data can also be processed, including data noise reduction, abnormal data identification and repair, and after normalization processing, a part of the traffic flow data (for example, the first 24 A total of 2016 traffic flow data per day) is used as historical data, and this part of historical data is reconstructed through phase space as training samples, that is, the wolf pack algorithm is used to train these historical data, and the initial wavelet neural network parameters are obtained, and the other part of data is (that is, 672 traffic flow data in the last 7 days) are used as test samples after phase space reconstruction (that is, as traffic flow data for prediction). That is, use the historical data of the first 24 days to train and initialize the wavelet neural network parameters to construct the wavelet neural network traffic flow forecasting model, and then implement single-point and single-step forecasting for the traffic flow of the next 7 days through the well-built wavelet neural network traffic flow forecasting model , to get the predicted result.

Of course, the above is just an example. In practical applications, historical data and forecast data can use the same set of historical traffic flow data, or different historical traffic flow data. The specific traffic flow data to be used as historical data and forecast data can be based on actual It depends on the situation, and the embodiment of the present invention does not make any special limitation thereto, as long as the purpose of the embodiment of the present invention can be achieved.

The embodiment of the present invention provides a traffic flow prediction method based on wolf pack algorithm, including: obtaining traffic flow data; using a pre-established wavelet neural network traffic flow prediction model to process the traffic flow data to obtain a traffic flow prediction result; wherein, The wavelet neural network traffic flow prediction model is trained based on the wolf pack algorithm. The training process is to calculate the initial wavelet neural network parameters based on historical data and wolf pack algorithm; use the wavelet neural network and historical data to initialize the wavelet neural network parameters. A wavelet neural network traffic flow prediction model is obtained through training.

It can be seen that the initial wavelet neural network parameters of the wavelet neural network traffic flow prediction model adopted in the embodiments of the present invention when predicting traffic flow are calculated based on historical data and the wolf pack algorithm. Because of its wide range, it can converge to the global optimal solution to a large extent. Therefore, when the wavelet neural network traffic flow prediction model trained by the initial wavelet neural network parameters based on the wolf pack algorithm is used to predict the traffic flow, To a certain extent, the prediction speed and prediction accuracy are improved.

The embodiment of the present invention discloses a traffic flow prediction method based on wolf pack algorithm. Compared with the previous embodiment, this embodiment further explains and optimizes the technical solution.

It should be noted that before training the wavelet neural network traffic flow prediction model, the length of the input traffic flow data or historical data and the control parameters of the model need to be set in advance. For example, the input length can be set as TFS (positive integer), then the traffic flow data (or traffic flow time series) whose input length is TFS is c={c(i)|i=1,2,,TFS};

The set control parameters may include: Input, that is, the number of neurons in the input layer; Hidden, that is, the number of neurons in the wavelet layer; Ouput, that is, the number of neurons in the output layer, where Input≤TFS.

In addition, it is also necessary to establish a wavelet neural network expressway traffic flow prediction model:

Among them, o represents the traffic flow output of the wavelet neural network; w _ij represents the connection weight of the i-th input and the j-th wavelet element; (F(1),F(2),,F(Input)) is the input traffic flow data (namely phase space reconstructed traffic flow input data); v _ij represents the weight connecting the wavelet layer and the output layer; b _j represents the jth translation coefficient; a _j represents the jth expansion coefficient; L represents the wavelet basis functions, and Among them, t is the time unit second. In the embodiment of the present invention, by calculating the initialization wavelet neural network parameters w _ij , v _ij , a _j , b _j (i=1,2,...,Input; j=1,2,...,Output), further The wavelet neural network traffic flow prediction model is obtained and used to predict traffic flow. specific:

In S21 of the previous embodiment, the process of calculating the initial wavelet neural network parameters based on historical data and wolf pack algorithm can be specifically:

S211: Encode each network parameter into each individual wolf according to the historical data, and the position of each individual wolf corresponds to each network parameter;

Specifically, encoding individual wolves is to encode each network parameter w _ij , v _ij , a _j , b _j (i=1,2,...,Input; j=1,2,...,Output), where the i-th The position of each D-dimensional individual wolf is x _i =(w _ij ,v _ij ,a _j ,b _j ) ^T , that is, the position of each individual wolf is in one-to-one correspondence with the corresponding network parameters, and D represents the wavelet The sum of the number of neural network parameters.

S212: Find the optimized position of the head wolf from each individual wolf according to the preset control parameters and corresponding strategies;

It should be noted that the preset control parameters are obtained by setting the control parameters in advance. The preset control parameters can include the total number of individual wolves N (that is, the total number of detective wolves, ferocious wolves, and alpha wolves), the maximum iteration The number of times Z _max , the maximum number of walks T _max , the preset number h (that is, the number of wolves detected) and the preset distance (that is, the minimum distance H _near between the ferocious wolf and the head wolf). You can also use T to represent the current number of walks; use Y _i to represent the current wolf's prey odor concentration (that is, the prey odor concentration detected by the current wolf); use Y _lead to represent the prey odor concentration of the head wolf; use z to represent the current algebraic iteration frequency.

Then, in the above S212, according to the preset control parameters and the corresponding strategies, the process of finding the optimized position of the head wolf from each individual wolf can be specifically as follows:

It should be noted that the walking strategy of the pack of wolves needs to be determined first, that is, a preset number of wolves other than the head wolf should be selected as scouting wolves, and the walking behavior improved by the optimization strategy of the law of universal gravitation should be implemented. Specifically, such as S2121-S2122:

S2121: Select a preset number of wolves other than the head wolf as wolf detectors, execute the walking behavior improved by the optimization strategy of the law of universal gravitation, and use the position corresponding to the wolf with the highest prey odor concentration in the current pack of wolves as the wolf detector wander. Walking direction; S2122: update the position of the wolf detection, and until the prey odor concentration of the detection wolf is greater than the prey odor concentration of the head wolf (ie Y _i >Y _lead ) or when the current number of travels T reaches the maximum number of travels T _max , the The position of the corresponding Detective Wolf replaces the position of the Head Wolf, and the Detective Wolf becomes the new Head Wolf;

Optionally, the preset control parameters in the embodiment of the present invention may include the total number of individual wolves, the maximum number of iterations, the maximum number of walks, the preset number, and the preset distance;

Of course, the preset control parameters are not limited to the above-mentioned parameters, but may also include other parameters, which may be determined according to the actual situation. The embodiment of the present invention does not specifically limit this, and the purpose of the embodiment of the present invention can be achieved. That's it.

Further, the process of updating the location of the wolf detection in S2122 may specifically be:

The position of the wolf detection is updated according to the first calculation relation; the first calculation relation is:

Among them, x′i represents the updated position of the _{i-th wolf detection; x i represents} the current position of the i-th detection wolf; x _k represents the position of the k-th detection wolf; The uniform distribution function of , x _best represents the position where the current prey odor concentration is the largest, and Y _ik represents the prey odor concentration function of the i-th wolf relative to the k-th wolf;

Y _ik is obtained according to the second calculation relation, which is: Among them, G represents the gravitational constant; Y(x _i ) represents the prey odor concentration of the i-th wolf detection (that is, the error between the training output value of the wavelet neural network and its expected value is the objective function value), Y(x _k ) represents the k-th , Only detect the concentration of prey odor of wolves.

It should be noted that the above Y( _xi ) can be expressed by the objective function Obtained, and the average value of the wolf detection position in the wolf pack satisfies Mode.

After determining the roaming strategy of the wolves, it is then necessary to determine the running strategy of the wolves, and call the wolves through the new head wolf generated above, so that the wolves execute the chaotic strategy to optimize the running behavior, as in S2123-S2124:

S2123: Summon the wolf through the new head wolf, and use the chaotic initialization to process the obtained wolf variable to obtain the position of the initialized wolf;

S2124: Generate the position of the new wolf individual according to the position of the initialized wolf, calculate the prey odor concentration of the new wolf individual, and replace the position of the new wolf individual when the prey odor concentration is greater than the prey odor concentration of the new head wolf The position of the head wolf; until the distance between the new wolf and the head wolf is less than the preset distance;

Further, in S2124, the process of generating the position of the new wolf individual according to the position of the initialized wolf can be specifically as follows:

The position of the new individual of the wolf is obtained according to the position of the initialized wolf and the third calculation relation. The third calculation relation is:

x′ _n =Y _i +R _i (2x _n,k -1), where x _n ′ is the position of the new wolf individual, and x _n,k represents the position Y _i of the new wolf individual at the kth iteration. The odor concentration of the prey; R _i is the radius of the attack area;

Wherein, x _{n, k} can be obtained according to the fourth calculation relation, and the fourth calculation relation is x _{n, k+1} = μ x n _{, k} (1-x _{n, k} ), wherein, n∈[1, N] , N represents the total number of individual wolves, μ represents the control parameters of the chaotic state, and k represents the number of iterations.

Specifically, the process is to map the variable x _n chaotically to the optimized variable x′ _n , where x′ _n runs with the prey concentration and smell Y _i as the center and R _i as the radius In the region, calculate the prey odor concentration Y(x′ _n ), and update the maximum prey odor concentration Y _best (x′ _n ) in the chaotic iteration process. If the current wolf’s prey odor concentration is greater than the prey odor concentration of the head wolf, then Replace the alpha wolf. Until the distance between the ferocious wolf and the alpha wolf is less than the preset distance H _near .

It should be noted that the value of μ in the embodiment of the present invention may be 4, and when μ is 4, the state of chaos may be completely entered. Of course, in practical applications, the value of μ is not limited to 4, but can also be other values. For example purposes.

After determining the attack strategy of the wolves, it is necessary to determine the siege strategy of the wolves and implement the siege behavior optimized by the inertia weight adaptive strategy, as in S2125:

S2125: Execute the siege behavior optimized by the inertia weight adaptive strategy, and update the position of the head wolf;

Further, in S2125, the siege behavior optimized by the inertial weight adaptive strategy is executed, and the process of updating the position of the head wolf can be specifically as follows:

Execute the siege behavior optimized by the inertia weight adaptive strategy, and update the position of the head wolf according to the fifth calculation relation, the fifth calculation relation is Among them, M ^k represents the position of the prey when the number of iterations is k, step _c represents the attack step size of the wolf, and γ represents the inertia weight; γ is obtained according to the sixth calculation relation, which is:

Among them, γ _max represents the maximum inertia weight; γ _min represents the minimum inertia weight, Z _max represents the maximum number of iterations, and z represents the current number of iterations.

S2126: Update the position of the wolf pack according to the "winner is king" rule, and then update the wolf pack according to the "survival of the strong" mechanism and the principle of "survival of the strong, the jungle of the jungle";

Specifically, when updating the wolf pack, the P wolves farthest from the prey (objective function value) can be eliminated and P wolves can be randomly generated to update the wolf pack. Among them, P∈[N/(2β),N/β], β is the update scaling factor.

S2127: Determine whether the preset accuracy is reached between the alpha wolf and the prey or the current number of iterations is greater than the maximum number of iterations, if yes, use the alpha wolf as the optimized alpha wolf, and output the position of the optimized alpha wolf; otherwise, return S2121.

S213: Decode the position to obtain a network parameter corresponding to the position, and use the network parameter as an initialization wavelet neural network parameter.

Specifically, since the position of each individual wolf in the wolf pack is in one-to-one correspondence with the network parameters, when the optimized alpha wolf is found, the initial wavelet neural network parameters can be obtained by decoding the optimized alpha wolf position.

It should be noted that after the initial wavelet neural network parameters are calculated, the wavelet neural network and historical data can be used to train the initial wavelet neural network parameters to obtain the wavelet neural network traffic flow prediction model. That is, the process of S22 is as follows:

S221: According to the Input neurons in the input layer, use the G-P algorithm to reconstruct the traffic flow sequence phase space (that is, input the Input historical data (that is, the traffic flow sequence) to predict the Input+1th traffic flow time sequence) to obtain training input samples and training output samples .

S222: Establish a training objective function Among them, E represents the mean square error function between the expected value of the wavelet neural network traffic flow prediction and the actual output value of the network; sp represents the number of training sample groups; s _j represents the output of the expected value of the jth traffic flow.

S223: If |E| is greater than the set value, if according to formula as well as Update the wavelet neural network parameters, skip to S222, to correct the wavelet neural network parameters; where η is the wavelet neural network learning factor.

Substituting the trained wavelet neural network parameters w _ij , v _ij , a _j , b _j (i=1,2,…,Input; j=1,2,…,Output) into the predicted wavelet neural network, the wavelet neural network can be obtained Network traffic flow forecasting model, and use the acquired traffic flow data and calculation relation Obtain the predicted output of the wavelet neural network.

In addition, please refer to FIG. 2 and FIG. 3. FIG. 2 is a schematic diagram of expressway traffic flow prediction simulation using the wolf pack algorithm-based traffic flow prediction method provided by the embodiment of the present invention. FIG. Schematic diagram of expressway traffic flow forecasting simulation of network traffic flow forecasting method. IWN-WNN in Fig. 2 represents the wavelet neural network prediction method based on wolf pack algorithm; WNN in Fig. 3 represents the prediction method based on wavelet neural network. It can be seen from FIG. 2 and FIG. 3 that the traffic flow prediction method based on the wolf pack algorithm provided by the embodiment of the present invention has higher accuracy and better prediction effect.

Correspondingly, the embodiment of the present invention discloses a traffic flow prediction device based on wolf pack algorithm. Please refer to FIG. 4 for details. FIG. 4 is a schematic structural diagram of a traffic flow prediction device based on wolf pack algorithm provided by an embodiment of the present invention. On the basis of above-mentioned embodiment:

The unit includes:

Obtaining module 1, used to obtain traffic flow data;

The processing module 2 is used to process the traffic flow data by using the pre-established wavelet neural network traffic flow prediction model to obtain the traffic flow prediction result; wherein, the wavelet neural network traffic flow prediction model includes:

Calculation module, used to calculate the initial wavelet neural network parameters according to historical data and wolf pack algorithm;

The training module is used to use the wavelet neural network and historical data to train the initial wavelet neural network parameters to obtain the wavelet neural network traffic flow prediction model.

It should be noted that the traffic flow prediction system based on the wolf pack algorithm provided by the embodiment of the present invention can improve the prediction speed and prediction accuracy to a certain extent during use.

In addition, for the specific introduction of the traffic flow prediction method based on the wolf pack algorithm involved in the embodiment of the present invention, please refer to the above method embodiment, and the present application will not repeat it here.

On the basis of the above embodiments, please refer to FIG. 5 , which is a schematic structural diagram of a wavelet neural network traffic flow prediction model provided by an embodiment of the present invention.

Optionally, computing modules include:

The individual wolf encoding unit is used to encode each network parameter into each individual wolf according to historical data, and the position of each individual wolf corresponds to each network parameter;

The head wolf finds the unit, which is used to find the optimized position of the head wolf from each individual wolf according to the preset control parameters and corresponding strategies;

The decoding unit is used to decode the position to obtain network parameters corresponding to the position, and use the network parameters as initial wavelet neural network parameters.

In order to solve the above-mentioned technical problems, an embodiment of the present invention provides a traffic flow prediction system based on wolf pack algorithm, including the above-mentioned traffic flow prediction device based on wolf pack algorithm.

It should be noted that the traffic flow prediction system based on the wolf pack algorithm provided by the embodiment of the present invention can improve the prediction speed and prediction accuracy to a certain extent during use.

In addition, for the specific introduction of the traffic flow prediction method based on the wolf pack algorithm involved in the embodiment of the present invention, please refer to the above method embodiment, and the present application will not repeat it here.

It should also be noted that in this specification, relative terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that these entities or operations There is no such actual relationship or order between the operations. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising said element.

The above description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (9)

1. a traffic flow forecasting method based on wolf pack algorithm, is characterized in that, described method comprises: Obtain traffic flow data; Using the pre-established wavelet neural network traffic flow prediction model to process the traffic flow data to obtain the traffic flow prediction result; wherein, the wavelet neural network traffic flow prediction model is trained based on the wolf pack algorithm, and the training process is as follows: : Calculate the initial wavelet neural network parameters based on historical data and wolf pack algorithm; The wavelet neural network traffic flow prediction model is obtained by using the wavelet neural network and the historical data to train the initialized wavelet neural network parameters. 2. the traffic flow prediction method based on wolf pack algorithm according to claim 1, is characterized in that, described according to historical data and wolf pack algorithm calculates the process that initializes wavelet neural network parameter to be specifically: encoding each network parameter into each individual wolf according to the historical data, and the position of each individual wolf corresponds to each network parameter; Finding the position of the optimized head wolf from each of the individual wolves according to preset control parameters and corresponding strategies; The position is decoded to obtain network parameters corresponding to the position, and the network parameters are used as initial wavelet neural network parameters. 3. the traffic flow prediction method based on wolf pack algorithm according to claim 2, is characterized in that, The preset control parameters include the total number of individual wolves, the maximum number of iterations, the maximum number of walks, the preset number and the preset distance; The process of finding the position of the optimized alpha wolf from each of the individual wolves according to the preset control parameters and corresponding strategies is specifically as follows: S2121: Select a preset number of wolves other than the head wolf as wolf detectors, execute the walking behavior improved by the optimization strategy of the law of universal gravitation, and use the position corresponding to the wolf with the highest prey odor concentration in the current pack of wolves as the wolf detector wander. walk direction; S2122: Update the position of the wolf detection, and until the prey odor concentration of the wolf detection is greater than the prey odor concentration of the head wolf or the current number of wandering reaches the maximum number of travels, replace the corresponding wolf detection position with that of the head wolf position, the detective wolf becomes the new head wolf; S2123: Summon ferocious wolves through the new alpha wolf, and process the acquired wolf variables by using chaotic initialization to obtain the position of the initialized ferocious wolf; S2124: Generate the position of the new wolf individual according to the position of the initialized wolf, calculate the prey odor concentration of the new wolf individual, and when the prey odor concentration is greater than the prey odor concentration of the new head wolf, calculate the The position of the new wolf individual replaces the position of the head wolf; until the distance between the new wolf individual and the head wolf is less than the preset distance; S2125: Execute the siege behavior optimized by the inertia weight adaptive strategy, and update the position of the head wolf; S2126: Update the position of the wolf pack according to the "winner is king" rule, and then update the wolf pack according to the "survival of the strong" mechanism and the principle of "survival of the strong, the jungle of the jungle"; S2127: Determine whether the preset accuracy is reached between the alpha wolf and the prey or the current number of iterations is greater than the maximum number of iterations, if so, use the alpha wolf as the optimized alpha wolf, and output the position of the optimized alpha wolf ; Otherwise, return to S2121. 4. the traffic flow prediction method based on wolf pack algorithm according to claim 3, is characterized in that, the process of described updating the position of wolf detection is specifically: The position of the wolf detection is updated according to the first calculation relation; the first calculation relation is: Wherein, the x′ _i represents the updated position of the i-th wolf detector; the x _i represents the current position of the i-th wolf detector, the x _k represents the position of the k-th wolf detector, and the rand(0 , 1) represents a uniform distribution function from 0 to 1, the x _best represents the position where the current prey odor concentration is the largest, and the Y _ik represents the prey odor concentration function of the i-th wolf relative to the k-th wolf; The Y _ik is obtained according to the second calculation relational expression, and the second calculation relational expression is: Wherein, the G represents the gravitational constant; the Y( _xi ) represents the prey odor concentration of the i-th wolf detection, and the Y(x _k ) represents the prey odor concentration of the k-th wolf detection. 5. the traffic flow prediction method based on wolf pack algorithm according to claim 3, is characterized in that, the process of generating the position of the new individual of fierce wolf according to the position of initialization fierce wolf is specifically: Obtain the position of the new individual of the wolf according to the position of the initialization wolf and the third calculation relation, and the third calculation relation is: x _n ′=Y _i +R _i (2x _n,k -1), wherein, the x′ _n is the position of the new wolf individual, and the x _n,k represents the position of the new wolf individual at the kth iteration Position; said Y _i is the odor concentration of the prey; said R _i is the radius of the attack area; The x _n,k is obtained according to the fourth calculation relational expression, and the fourth calculation relational expression is x _n,k+1 =μx _n,k (1-x _n,k ), wherein, n∈[1,N ], the N represents the total number of individual wolves, the μ represents the control parameters of the chaotic state, and the k represents the number of iterations. 6. the traffic flow prediction method based on wolf pack algorithm according to claim 3, is characterized in that, the siege behavior of described execution inertial weight self-adaptive strategy optimization, and the process that the position of head wolf is updated is specifically: Execute the siege behavior optimized by the inertia weight adaptive strategy, and update the position of the head wolf according to the fifth calculation relation, the fifth calculation relation is Wherein, said M ^k represents the position of the prey when the number of iterations is k, said step _c represents the attack step size of the wolf, and said γ represents the inertia weight; said γ is obtained according to the sixth calculation relational formula, and the sixth The calculation relation is: Wherein, the γ _max represents the maximum inertial weight; the γ _min represents the minimum inertial weight, the Z _max represents the maximum number of iterations, and the z represents the current number of iterations. 7. A traffic flow prediction device based on wolf pack algorithm, characterized in that, said device comprises: An acquisition module, configured to acquire traffic flow data; A processing module, configured to process the traffic flow data using a pre-established wavelet neural network traffic flow prediction model to obtain a traffic flow prediction result; wherein the wavelet neural network traffic flow prediction model includes: Calculation module, used to calculate the initial wavelet neural network parameters according to historical data and wolf pack algorithm; The training module is configured to use the wavelet neural network and the historical data to train the parameters of the initialized wavelet neural network to obtain the traffic flow prediction model of the wavelet neural network. 8. the traffic flow forecasting device based on wolf pack algorithm according to claim 7, is characterized in that, described calculating module comprises: An individual wolf encoding unit, configured to encode each network parameter into each individual wolf according to historical data, and the position of each individual wolf is in one-to-one correspondence with each of the network parameters; The head wolf finding unit is used to find the optimized position of the head wolf from each of the individual wolves according to preset control parameters and corresponding strategies; A decoding unit, configured to decode the position to obtain a network parameter corresponding to the position, and use the network parameter as an initialization wavelet neural network parameter. 9. A traffic flow forecasting system based on wolf pack algorithm, characterized in that it comprises the traffic flow forecasting device based on wolf pack algorithm according to claim 7 or 8.