A freight vehicle and cargo matching information processing system

Abstract

The present invention relates to the field of information processing technology, and particularly to a matching information processing system for freight vehicles and goods. The content includes: collecting and preprocessing vehicle and goods data, extracting features and reducing dimensions based on the preprocessed data; obtaining the features of vehicles and goods from the data after dimension reduction, and combining multi-level weight adjustment and non-linear optimization to obtain the final matching result; generating a vehicle scheduling plan based on the final matching result, and sending the scheduling information to freight drivers and customers. It solves the problems that it is difficult to effectively process the data of complex and changeable transportation environments, the non-linear characteristics of vehicle and goods data are not fully considered, resulting in the inability of the data to comprehensively reflect the internal structure of the original data after dimension reduction; the lack of a dynamic adjustment mechanism, unable to perform real-time optimization according to the actual transportation situation; and the neglect of intelligence and real-time nature, the scheduling information is lagging and lacks flexibility, and cannot meet the complex and changeable transportation requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of information processing, and in particular to a system for processing matching information between freight vehicles and goods. Background Art

[0002] The freight logistics industry is an important part of the modern economy, and its efficient operation has a significant impact on social production and life. With the rapid development of e-commerce and the increasing frequency of global trade, the demand for freight transportation is constantly increasing. The freight logistics industry is facing huge challenges and opportunities, and the traditional freight management methods can no longer meet the requirements of high efficiency, accuracy, and intelligence in modern logistics. Therefore, how to use information processing technology and intelligent algorithms to improve the efficiency and accuracy of matching freight vehicles and goods has become an important problem to be solved urgently.

[0003] In the process of freight transportation, the matching of vehicles and goods is one of the core links. Accurate and efficient matching of vehicles and goods can significantly improve transportation efficiency, reduce transportation costs, and enhance customer satisfaction. With the development of technologies such as the Internet of Things, cloud computing, and big data, the freight logistics industry is gradually moving towards the direction of intelligence and informatization. The widespread application of in-vehicle GPS, Internet of Things sensors, and handheld terminal devices makes it possible to collect and transmit relevant data of vehicles and goods in real time. At the same time, the progress of artificial intelligence and machine learning technologies provides strong support for large-scale data processing and intelligent decision-making.

[0004] The existing technologies have at least the following technical problems: it is difficult to effectively process the data of complex and changeable transportation environments, and the non-linear characteristics of vehicle and goods data are not fully considered, resulting in the inability of the data to fully reflect the internal structure of the original data after dimensionality reduction processing, affecting the effect of the matching algorithm, and the matching results are often not accurate and reasonable enough; there is a lack of a dynamic adjustment mechanism and it is unable to perform real-time optimization according to the actual transportation situation, easily leading to the mismatch between the matching results and the actual needs, ignoring intelligence and real-time nature, the scheduling information is lagging and lacks flexibility, and it cannot meet the complex and changeable transportation requirements. Summary of the Invention

[0005] The present invention provides a system for processing matching information between freight vehicles and goods to solve the problems that it is difficult to effectively process the data of complex and changeable transportation environments, the non-linear characteristics of vehicle and goods data are not fully considered, resulting in the inability of the data to fully reflect the internal structure of the original data after dimensionality reduction processing, affecting the effect of the matching algorithm, and the matching results are often not accurate and reasonable enough; there is a lack of a dynamic adjustment mechanism and it is unable to perform real-time optimization according to the actual transportation situation, easily leading to the mismatch between the matching results and the actual needs, ignoring intelligence and real-time nature, the scheduling information is lagging and lacks flexibility, and it cannot meet the complex and changeable transportation requirements.

[0006] A freight vehicle and cargo matching information processing system of the present invention specifically includes the following technical solutions:

[0007] A method for processing freight vehicle and cargo matching information includes the following steps:

[0008] S1. Collect and preprocess vehicle and cargo data, extract features based on the preprocessed data, perform data dimensionality reduction on the extracted data features through a multi-layer non-linear dimensionality reduction method to obtain the dimensionality-reduced data;

[0009] S2. Obtain the features of the vehicle and cargo from the dimensionality-reduced data, combine multi-level weight adjustment and non-linear optimization, and obtain the final matching result through feedback adjustment and multi-objective balancing method; based on the final matching result, generate a vehicle scheduling plan and send the scheduling information to the freight driver and the customer.

[0010] Preferably, the S1 specifically includes:

[0011] Based on the preprocessed data, adopt an adaptive multi-scale feature extraction method, combine Fourier transform and wavelet transform for feature extraction to obtain the extracted data features.

[0012] Preferably, the S1 specifically includes:

[0013] Based on the extracted data features, design a multi-layer non-linear dimensionality reduction method, combine manifold learning and principal component analysis method, and map the high-dimensional data to the low-dimensional space through non-linear mapping to obtain the dimensionality-reduced data.

[0014] Preferably, the S2 specifically includes:

[0015] Obtain the features of the vehicle and cargo from the dimensionality-reduced data, define the fitness function of the initial matching, and calculate the initial matching result.

[0016] Preferably, the S2 specifically includes:

[0017] Based on the initial matching result, introduce time factor and vehicle type factor for weight adjustment to obtain the adjusted matching result.

[0018] Preferably, the S2 specifically includes:

[0019] Perform non-linear optimization on the adjusted matching result, introduce a regularization term, define the non-linear optimization objective function, and calculate the optimized result.

[0020] Preferably, the S2 specifically includes:

[0021] Dynamically adjust the matching result based on the optimized result, combine the historical matching result, and optimize to obtain the feedback-adjusted matching result.

[0022] Preferably, the S2 specifically includes:

[0023] Introduce a multi-objective balancing method, and obtain the final matching result by balancing the initial matching result, the adjusted matching result, and the feedback-adjusted matching result.

[0024] A freight vehicle and cargo matching information processing system includes the following parts:

[0025] A data acquisition module, a data processing module, an initial matching module, a weight adjustment module, a non-linear optimization module, a dynamic feedback adjustment module, a matching output module, and a scheduling module;

[0026] The data acquisition module collects vehicle and cargo data in real time; transmits the vehicle and cargo data to the data processing module through a wireless network;

[0027] The data processing module preprocesses the vehicle and cargo data, extracts features and reduces the data dimension based on the preprocessed data, and transmits the dimension-reduced data to the initial matching module;

[0028] The initial matching module obtains various features of the vehicle and the cargo from the dimension-reduced data, defines the fitness function of the initial matching, and obtains the initial matching result; transmits the initial matching result to the weight adjustment module and the matching output module;

[0029] The weight adjustment module introduces time factors and vehicle type factors for weight adjustment based on the initial matching result, further optimizes the matching result, and calculates the matching degree between the vehicle and the cargo to obtain the adjusted matching result; transmits the adjusted matching result to the non-linear optimization module and the matching output module;

[0030] The non-linear optimization module performs non-linear optimization on the adjusted matching result and comprehensively balances various indicators; updates the feature weights using the gradient descent method until the non-linear optimization objective function converges to obtain the optimized result; transmits the optimized result to the dynamic feedback adjustment module;

[0031] The dynamic feedback adjustment module dynamically adjusts the matching result based on the optimized result, combines the historical matching result, and optimizes to obtain the feedback-adjusted matching result; transmits the feedback-adjusted matching result to the matching output module;

[0032] The matching output module obtains the final matching result by balancing the initial matching result, the adjusted matching result, and the feedback-adjusted matching result; transmits the final matching result to the scheduling module;

[0033] The scheduling module generates a vehicle scheduling plan based on the final matching result and sends the scheduling information to the freight driver and the customer.

[0034] The beneficial effects of the technical solution of the present invention are as follows:

[0035] 1. By combining the advantages of Fourier transform and wavelet transform, the optimal features are adaptively selected to extract important features from vehicle and cargo data, effectively process complex multi-scale data, improve the effect of feature extraction, and provide high-quality input data for subsequent dimensionality reduction and matching; by combining manifold learning and principal component analysis method, the high-dimensional data is mapped to a low-dimensional space through non-linear mapping, effectively reducing the dimension of the data, improving the efficiency of data processing and matching, while retaining the key features of vehicle and cargo data, ensuring the accuracy and reliability of the matching result.

[0036] 2. In the initial matching stage, by considering factors such as distance, weight, volume and time window, the initial matching of vehicles and cargoes is realized; subsequently, in the weight adjustment stage, the time factor and vehicle type factor are further introduced to adjust and optimize the initial matching result; this multi-level matching method can comprehensively consider various factors and improve the accuracy and rationality of matching.

[0037] 3. In the non-linear optimization stage, by introducing a regularization term, overfitting is prevented, ensuring that the matching process has good generalization ability when dealing with new data; the design of the non-linear optimization objective function can achieve the optimal matching effect within the global range, effectively avoiding the problem of local optimal solutions, improving the global optimality and robustness of the matching result; by combining historical matching results, the matching result is further optimized; the dynamic feedback adjustment mechanism can continuously optimize and adjust the matching result according to the actual situation, improving the adaptability and matching effect of the information processing system.

[0038] 4. In the final matching stage, by balancing the initial matching result, the adjusted matching result and the feedback-adjusted matching result, global optimization is realized. The multi-objective balancing method can comprehensively consider the matching results of different stages, ensuring the optimality and rationality of the final matching result; a vehicle scheduling plan is generated according to the final matching result, and the scheduling information is sent to the freight driver and the customer through the scheduling module to ensure the best scheduling of vehicles and cargoes. The scheduling plan includes information such as vehicle routes and estimated arrival times, ensuring the efficiency and punctuality of the freight process. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a structural diagram of a freight vehicle and cargo matching information processing system according to the present invention;

[0040] Figure 2 It is a flowchart of a freight vehicle and cargo matching information processing method according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0041] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0043] The following specifically describes the specific solution of a freight vehicle and cargo matching information processing system provided by the present invention in conjunction with the accompanying drawings.

[0044] Refer to the attached Figure 1 , which shows the structure diagram of a freight vehicle and cargo matching information processing system provided by an embodiment of the present invention. The system includes the following parts:

[0045] Data acquisition module, data processing module, initial matching module, weight adjustment module, non-linear optimization module, dynamic feedback adjustment module, matching output module, scheduling module;

[0046] The data acquisition module, through in-vehicle GPS, Internet of Things sensors and handheld terminal devices, real-time collects vehicle and cargo data; transmits the vehicle and cargo data to the data processing module through a wireless network;

[0047] The data processing module preprocesses the collected vehicle and cargo data, including data cleaning and outlier detection, to obtain preprocessed data; performs feature extraction and data dimensionality reduction on the preprocessed data to obtain reduced-dimensional data; transmits the reduced-dimensional data to the initial matching module;

[0048] The initial matching module obtains various features of vehicles and cargoes from the reduced-dimensional data, defines the fitness function for initial matching, and obtains the initial matching result; transmits the initial matching result to the weight adjustment module and the matching output module;

[0049] The weight adjustment module, based on the initial matching result, introduces time factors and vehicle type factors for weight adjustment to further optimize the matching result. By comparing the type characteristics of the vehicle with the demand characteristics of the cargo, calculates the matching degree between the vehicle and the cargo, comprehensively considers distance, weight, volume, time and vehicle type, improves the accuracy of matching, and obtains the adjusted matching result; transmits the adjusted matching result to the non-linear optimization module and the matching output module;

[0050] The non-linear optimization module performs non-linear optimization on the adjusted matching results, introduces a regularization term, defines a non-linear optimization objective function, conducts global optimization processing to avoid the problem of local optimal solutions, and further comprehensively balances various indicators to achieve the optimal matching effect within the global scope. The gradient descent method is used to update the feature weights until the non-linear optimization objective function converges to obtain the optimized results. The optimized results are transmitted to the dynamic feedback adjustment module.

[0051] The dynamic feedback adjustment module dynamically adjusts the matching results based on the optimized results, combines the historical matching results, and further optimizes to obtain the matching results after feedback adjustment. The matching results after feedback adjustment are transmitted to the matching output module.

[0052] The matching output module achieves global optimization by balancing the initial matching results, the adjusted matching results, and the matching results after feedback adjustment to obtain the final matching results. The final matching results are transmitted to the scheduling module.

[0053] The scheduling module generates a vehicle scheduling plan based on the final matching results and sends the scheduling information to the freight driver and the customer.

[0054] Refer to Appendix Figure 2 , which shows a flowchart of a method for processing matching information between freight vehicles and goods provided by an embodiment of the present invention. The method includes the following steps:

[0055] S1. Collect and preprocess vehicle and cargo data, extract features based on the preprocessed data, and perform data dimensionality reduction on the extracted data features through a multi-layer non-linear dimensionality reduction method to obtain the dimensionality-reduced data.

[0056] The freight vehicle and cargo matching information processing system collects vehicle and cargo data in real time through in-vehicle GPS, Internet of Things sensors, and handheld terminal devices. Among them, vehicle data includes vehicle location, load capacity, current load, vehicle type, fuel status, driver status, etc.; cargo data includes cargo location, cargo type, cargo weight, destination, delivery time window, etc. The vehicle and cargo data are transmitted to the data processing module through a wireless network.

[0057] The data processing module includes multiple stages. Since the collected vehicle and cargo data may contain noise and error values, the data processing module first needs to preprocess the collected vehicle and cargo data, including data cleaning and outlier detection, to remove noise and outliers and ensure the accuracy and reliability of the data, obtaining the preprocessed data. Next, feature extraction and data dimensionality reduction are performed on the preprocessed data. An adaptive multi-scale feature extraction method is used, combining the advantages of Fourier transform and wavelet transform, to adaptively select the best features.

[0058] The formula for adaptive multi-scale feature extraction is as follows:

[0059]

[0060] Among them, X(t) is the data feature extracted at time t; x(t) is the preprocessed data; ψ k is the wavelet function of the k-th scale; K is the number of types of wavelet functions; a k is the scale parameter of the wavelet function of the k-th scale; τ k is the translation parameter of the wavelet function of the k-th scale; ω k is the frequency parameter of the Fourier transform; is the imaginary unit. By adaptively selecting different scales and frequency components, important features in vehicle and cargo data are extracted.

[0061] Design a multi-layer non-linear dimensionality reduction method. Input the extracted data features, combine manifold learning and principal component analysis method, and map high-dimensional data to a low-dimensional space through non-linear mapping, retaining the main features of vehicle and cargo data; Manifold learning aims to perform dimensionality reduction by learning the low-dimensional manifold structure of data. Common manifold learning methods include locally linear embedding and Laplacian operator. Taking locally linear embedding as an example, the specific implementation process of the manifold learning method is as follows:

[0062] For each vehicle and cargo data feature, find its k nearest neighbors; Let X u be the u-th data feature; represents the neighbor set of X u , calculate the weight matrix W between each data feature X u and its neighbors, so that X u can be linearly reconstructed by its neighbors; The calculation formula of the weight matrix W is as follows:

[0063]

[0064] Among them, X v is the neighbor data feature of X u , the constraint condition is ∑ v W uv = 1; W uv represents the weight between the data feature X u and X v .

[0065] By solving the following minimum eigenvalue problem, find the representation Φ(X u ) of the data feature X u in the low-dimensional space:

[0066]

[0067] Among them, Φ(X u ) represents the representation of X u in the low-dimensional space, that is, the low-dimensional data of X u .

[0068] Based on manifold learning, the principal component analysis method is used to further process the embedded low-dimensional data;

[0069] First, decentralize the low-dimensional data Φ(X) to obtain the decentralized data

[0070]

[0071] where μ Φ(X) is the mean vector of Φ(X);

[0072] Then, calculate the covariance matrix of the decentralized data

[0073]

[0074] where U is the number of decentralized data; T represents the transpose of the matrix;

[0075] Furthermore, perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalue Λ and the eigenvector matrix E:

[0076]

[0077] Select the eigenvectors corresponding to the first N largest eigenvalues to form the projection matrix Finally, the data after dimensionality reduction is expressed as:

[0078]

[0079] The data z after dimensionality reduction includes various characteristics of the vehicle and the goods, such as vehicle location, load capacity, current load, vehicle type, fuel status, driver status, goods location, goods type, goods weight, destination, and delivery time window; the data after dimensionality reduction is transmitted to the initial matching module.

[0080] S2. Obtain the characteristics of the vehicle and the goods from the data after dimensionality reduction, combine multi-level weight adjustment and non-linear optimization, and through feedback adjustment and multi-objective balance method, obtain the final matching result; based on the final matching result, generate a vehicle scheduling plan and send the scheduling information to the freight driver and the customer.

[0081] Combine multi-level weight adjustment and non-linear optimization to achieve the best matching of freight vehicles and goods through a multi-level structure;

[0082] In the initial matching module, various features of the vehicle and the goods are obtained from the dimensionality-reduced data, and rough matching is performed using the initial rules.

[0083] There are a vehicle set V and a goods set C, and the fitness function of the initial matching is defined as:

[0084]

[0085] where f init (V i , C j ) is the fitness function of the initial matching, indicating the initial matching result of vehicle V i and goods C j ; d ij represents the distance between vehicle V i and goods C j ; wd j represents the weight of goods C j ; WD i represents the maximum load capacity of vehicle V i ; e j represents the volume of goods C j ; E i represents the maximum load volume of vehicle V i ; f j represents the urgency of the delivery time window of goods C j ; F i represents the schedulable time of vehicle V i ; By considering distance, weight, volume, and time window, preliminary vehicle and goods matching is achieved.

[0086] The initial matching result f init (V i , C j ) is input into the weight adjustment module, and time factor and vehicle type factor are further introduced for weight adjustment. The adjusted fitness function is:

[0087]

[0088] where f adjust (V i , C j ) is the adjusted fitness function, indicating the adjusted matching result of vehicle V i and goods C j ; α is the weight coefficient of the initial matching result; t ij represents the estimated transportation time between vehicle V i and goods C j ; t max represents the maximum allowable transportation time; β is the weight coefficient of the time factor; T irepresents the vehicle type matching degree; γ is the weight coefficient of the vehicle type factor. The vehicle type matching degree T i is calculated by comparing the vehicle type characteristics and the cargo demand characteristics. First, obtain the relevant type characteristic information of vehicle V i , including the vehicle's size, load capacity, fuel type, refrigeration function, etc., and obtain the relevant demand characteristic information of cargo C j , including the cargo's size, weight, requirements for the transportation environment (such as temperature, humidity), transportation time limit, etc.; secondly, by comparing the vehicle type characteristics with the cargo demand characteristics, calculate the matching degree between vehicle V i and cargo C j . The specific method is to compare each type characteristic with the demand characteristic and calculate their matching degree. For example, for each characteristic, calculate the product of the characteristic value of the vehicle and the characteristic value of the cargo demand, then normalize it, and finally sum the matching degrees of all characteristics to obtain the total matching degree T i . By comprehensively considering distance, weight, volume, time, and vehicle type, the accuracy of the matching is improved.

[0089] Input the adjusted matching result f adjust (V i , C j ) into the non-linear optimization module to further improve the global optimality and robustness of the matching result. Although the multi-level weight adjustment has considered various factors such as distance, weight, time, and vehicle type, it only achieves the optimal matching within a local range; the non-linear optimization module can effectively avoid the problem of local optimal solutions and further comprehensively balance various indicators through global optimization processing of the adjusted matching result, so as to achieve the optimal matching effect within the global range; the optimization process of the non-linear optimization module also introduces a regularization term to prevent overfitting and ensure that the matching process has good generalization ability when dealing with new data. The non-linear optimization objective function is defined as:

[0090]

[0091] where J(z) is the non-linear optimization objective function; λ is the regularization parameter; z u represents the weight of the u-th feature; N is the number of feature weights; n and m are the numbers of vehicles and cargos respectively. By minimizing the non-linear optimization objective function, global optimization processing is carried out and overfitting is prevented.

[0092] Derive the non-linear optimization objective function to calculate the gradient:

[0093]

[0094] Use the gradient descent method to update the feature weight z u :

[0095]

[0096] Among them, η is the learning rate; z u (k) represents the feature weight at the k-th iteration; z u (k + 1) represents the updated feature weight. Repeat the calculation of the gradient and the gradient descent step until the non-linear optimization objective function J(z) converges to obtain the optimized result f opt (V i , C j ).

[0097] Input the optimized result f opt (V i , C j ) into the dynamic feedback adjustment module, and dynamically adjust the matching result according to the optimized result f opt (V i , C j ). The fitness function after feedback adjustment is defined as:

[0098] f fb (V i , C j ) = f opt (V i , C j ) + θ · R ij

[0099] Among them, f fb (V i , C j ) is the fitness function after feedback adjustment, representing the matching result after feedback adjustment between vehicle V i and cargo C j ; R ij represents the historical matching result between vehicle V i and cargo C j , which is from the existing database; θ is the feedback adjustment coefficient. By combining the historical matching result, the matching result is further optimized.

[0100] The matching output module uses the multi-objective balance method to balance each objective function and achieve global optimization. The multi-objective balance function is defined as:

[0101] f final (V i , C j ) = λ 1 · f init (V i , C j ) + λ 2 · f adjust (V i , Cj ) + λ 3 ·f fb (V i , C j )

[0102] where f final (V i , C j ) is the final matching result of vehicle V i and cargo C j ; λ 1 , λ 2 , λ 3 are balance coefficients. By balancing the initial matching result, the adjusted matching result, and the feedback-adjusted matching result, the final matching result is obtained.

[0103] The scheduling module generates a vehicle scheduling plan based on the final matching result and sends the scheduling information to the freight driver and the customer. The scheduling information includes the vehicle route, the estimated arrival time, etc. The scheduling information is generated from the output of the multi-objective balancing function to ensure the optimal scheduling of vehicles and cargo.

[0104] In summary, a matching information processing system for freight vehicles and cargo is completed.

[0105] The order of the invention embodiments is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0106] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments.

[0107] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for processing information on matching freight vehicles and goods, characterized in that: The following steps are involved: S1. Collect and preprocess vehicle and cargo data, perform feature extraction based on the preprocessed data, and perform data dimension reduction based on the extracted data features using a multi-layer nonlinear dimension reduction method to obtain dimension-reduced data; S2. Obtain the characteristics of vehicles and goods from the data after dimensionality reduction. Suppose there is a vehicle set V and a cargo set C, define the fitness function of the initial matching, and calculate the initial matching result; the fitness function of the initial matching is defined as: Among them, f init (V i ,C j ) is the fitness function of the initial matching, indicating that the vehicle V i With cargo C j The initial matching result of ij Indicates vehicle V i With cargo C j The distance between j Indicates Goods C j Weight; WD i Indicates vehicle V i Maximum load; e j Indicates Goods C j The volume of E i Indicates vehicle V i The maximum load volume; f j Indicates Goods C j The urgency of the delivery time window; F i Indicates vehicle V i The dispatchable time; Based on the initial matching results, the time factor and vehicle type factor are introduced to adjust the weights to obtain the adjusted matching results. The specific formula is: Among them, f adjust (V i ,C j ) is the adjusted fitness function, indicating that the vehicle V i With cargo C j The adjusted matching result; α is the weight coefficient of the initial matching result; t ij Indicates vehicle V i With cargo C j Estimated shipping time between max represents the maximum permissible transportation time; β is the weight coefficient of the time factor; T i represents the vehicle type matching degree; γ is the weight coefficient of the vehicle type factor; The adjusted matching results are nonlinearly optimized, regularization terms are introduced, nonlinear optimization objective functions are defined, and the optimized results are calculated; the nonlinear optimization objective function is defined as: Among them, J(z) is the nonlinear optimization objective function; λ is the regularization parameter; z u represents the u-th feature weight; N is the number of feature weights; n and m are the number of vehicles and goods respectively; The matching results are dynamically adjusted based on the optimized results, and the matching results after feedback adjustment are optimized and combined with the historical matching results. A multi-objective balancing method is introduced to obtain the final matching results by balancing the initial matching results, the adjusted matching results and the feedback adjusted matching results. Based on the final matching results, a vehicle dispatch plan is generated, and the dispatch information is sent to freight drivers and customers.

2. The method for processing freight vehicle and cargo matching information according to claim 1, characterized in that: The S1 specifically includes: Based on the preprocessed data, an adaptive multi-scale feature extraction method is used to extract features by combining Fourier transform and wavelet transform to obtain the extracted data features.

3. The method for processing freight vehicle and cargo matching information according to claim 2, characterized in that: The S1 specifically includes: Based on the extracted data features, a multi-layer nonlinear dimensionality reduction method is designed. Combining manifold learning and principal component analysis, high-dimensional data is mapped to low-dimensional space through nonlinear mapping to obtain the reduced-dimensional data.

4. A freight vehicle and cargo matching information processing system, applied to the freight vehicle and cargo matching information processing method according to claim 1, characterized in that: Includes the following parts: Data acquisition module, data processing module, initial matching module, weight adjustment module, nonlinear optimization module, dynamic feedback adjustment module, matching output module, scheduling module; The data acquisition module collects vehicle and cargo data in real time and transmits the vehicle and cargo data to the data processing module via a wireless network; The data processing module pre-processes the vehicle and cargo data, performs feature extraction and data dimension reduction based on the pre-processed data, and transmits the dimension-reduced data to the initial matching module; The initial matching module obtains various characteristics of vehicles and goods from the dimension-reduced data, defines the fitness function of the initial matching, and obtains the initial matching results; the initial matching results are transmitted to the weight adjustment module and the matching output module; The weight adjustment module introduces time factors and vehicle type factors to adjust the weights based on the initial matching results, further optimizes the matching results, and calculates the matching degree between the vehicle and the goods to obtain the adjusted matching results; transmitting the adjusted matching results to the nonlinear optimization module and the matching output module; Nonlinear optimization module, which performs nonlinear optimization on the adjusted matching results and comprehensively balances various indicators; Use the gradient descent method to update the feature weights until the nonlinear optimization objective function converges to obtain the optimized result; transmit the optimized result to the dynamic feedback adjustment module; Dynamic feedback adjustment module dynamically adjusts the matching results based on the optimized results, and optimizes the matching results after feedback adjustment based on the historical matching results; Transmitting the feedback-adjusted matching result to a matching output module; A matching output module obtains a final matching result by balancing the initial matching result, the adjusted matching result and feeding back the adjusted matching result; The final matching result is transmitted to the scheduling module; The dispatch module generates a vehicle dispatch plan based on the final matching results and sends the dispatch information to freight drivers and customers.

Images