A digital integrated application system based on intelligent road management

Abstract

The application discloses a kind of digital integrated application systems based on highway wisdom management and maintenance, including data acquisition and sensing module, data transmission and communication module, digital integrated application module, highway wisdom management and maintenance module, data storage and management module, user monitoring and decision module, data acquisition and sensing module are used to collect highway data, data transmission and communication module are used for data transmission and communication, digital integrated application module is used for data fusion processing and analysis, highway wisdom management and maintenance module is used for highway state detection and preventive maintenance, data storage and management module is used for data storage and management, user monitoring and decision module is used to provide user interface and decision support, a kind of digital integrated application systems based on highway wisdom management and maintenance of the application, highway information extraction algorithm based on multi-modal data fusion is proposed to carry out multi-modal data fusion and information extraction, highway maintenance strategy optimization algorithm based on genetic algorithm is proposed to formulate and optimize highway maintenance strategy.

Description

TECHNICAL FIELD

[0001] The application relates to the fields of multi-modal data fusion, intelligent information extraction and genetic algorithm, and particularly relates to a digital integrated application system based on intelligent road management and maintenance. BACKGROUND

[0002] The multi-modal data fusion and intelligent information extraction technology is a multi-modal data processing and information extraction method, which aims to solve the problems of multi-modal data processing and information extraction in a sensor network. The multi-modal data fusion technology fuses and processes multi-modal data from different sensors, and the intelligent information extraction technology can intelligently extract information most valuable for road state monitoring and maintenance. Through the effective combination of the two technologies, the system can more comprehensively and accurately grasp the real-time state of the road, and provide a solid data foundation for intelligent road management and maintenance.

[0003] The genetic algorithm is an optimization algorithm based on evolutionary theory, which aims to solve the problems of road maintenance demand prediction and maintenance strategy optimization. By analyzing historical data and current state indicators, the genetic algorithm can predict future road maintenance needs and develop and optimize road maintenance strategies. The use of the genetic algorithm can significantly improve maintenance effectiveness and resource utilization efficiency, ensure the optimal operation of the road, prolong its service life, and reduce maintenance costs.

[0004] However, the existing digital integrated application system based on intelligent road management and maintenance lacks effective methods for multi-modal data processing and information extraction, resulting in low data utilization efficiency and incomplete information extraction. In addition, the system lacks advanced algorithm support for road maintenance demand prediction and maintenance strategy optimization, which cannot accurately predict future maintenance needs, resulting in unscientific maintenance strategies and low resource utilization efficiency. SUMMARY

[0005] The application aims to provide a digital integrated application system based on intelligent road management and maintenance to solve the problems of the existing digital integrated application system based on intelligent road management and maintenance, such as lack of effective methods for multi-modal data processing and information extraction, resulting in low data utilization efficiency and incomplete information extraction, and lack of advanced algorithm support for road maintenance demand prediction and maintenance strategy optimization, which cannot accurately predict future maintenance needs, resulting in unscientific maintenance strategies and low resource utilization efficiency.

[0006] In order to achieve the above object, the present application provides the following technical scheme: a digital integrated application system based on intelligent management and maintenance of highways, comprising a data acquisition and sensing module, a data transmission and communication module, a digital integrated application module, a highway intelligent management and maintenance module, a data storage and management module, and a user monitoring and decision-making module, characterized in that: the data acquisition and sensing module is used to collect the state data of the highway in real time through a sensor network, including temperature, humidity, pressure and vibration information; the data transmission and communication module is used to transmit the collected data and communicate with the central server through wireless network and 5G technology; the digital integrated application module comprises a data fusion processing unit and an intelligent analysis and application unit; the data fusion processing unit proposes a highway information extraction algorithm based on multi-modal data fusion for data fusion processing of multi-modal data from the sensor network and intelligent extraction of effective information; the intelligent analysis and application unit is used to intelligently analyze the effective information extracted from the fused data and apply the analysis results to highway state prediction, maintenance strategy optimization and decision support; the highway intelligent management and maintenance module comprises a highway state monitoring unit and a maintenance strategy optimization unit; the highway state monitoring unit is used to monitor the specific state indicators of the highway in real time, including pavement temperature, humidity, crack width, settlement height and traffic flow; the maintenance strategy optimization unit proposes a highway maintenance strategy optimization algorithm based on genetic algorithm for formulating and optimizing the maintenance strategy of the highway to improve the maintenance effect and resource utilization efficiency; the data storage and management module is used to centrally store and manage the collected data and processed analysis data, ensuring the security and availability of the data; and the user monitoring and decision-making module is used to provide real-time monitoring interface and decision support function for users, including data visualization and early warning notification and decision control function.

[0007] Preferably, the data acquisition and sensing module collects multi-modal data of the highway in real time through the sensor network constructed on the highway, including temperature sensors, humidity sensors, pressure sensors and vibration sensors, to ensure that the state information of the highway can be obtained in real time, providing a basis for subsequent data processing and analysis.

[0008] Preferably, the data transmission and communication module uses wireless network and 5G communication technology to transmit the collected data to the central server in a timely manner, ensuring the real-time and accuracy of the data to support immediate decision-making and reaction.

[0009] Preferably, the digital integrated application module comprises a data fusion processing unit, which proposes a highway information extraction algorithm based on multi-modal data fusion, integrates and processes multi-modal data from different sensors to ensure the consistency and integrity of the data, and extracts effective information and utilization value of the multi-modal data.

[0010] Specifically, the highway information extraction algorithm based on multi-modal data fusion is as follows: first, assume the organization structure of a data unit, the data unit constitutes the basic unit of the overall multi-modal data fusion algorithm, each data unit contains an information core, a strategy set and a data pipeline set corresponding to the strategy, and the organization structure of the data unit is specifically represented by the formula:

[0011] U={C,S,O}

[0012] Wherein, U represents a data unit, C represents an information core, S represents a strategy set, and O represents a data pipeline set. The information core C is the core component of the data unit, carries the key information of the data intelligent implementation, the strategy set S includes the input and output strategy, the data standardization strategy and the mathematical logic operation strategy, the data pipeline O is the channel of information transmission and communication, and is one-to-one corresponding to the strategy. The association of the data pipeline generates a new information core, realizes the hierarchical structure of the data unit, then, the initial data unit is directly associated with the database, including the construction of the initial information core, the construction of the initial strategy and the construction of the initial data pipeline. The initial information core is the field name and technical name extracted from the database, which is the common attribute of data. The initial strategy set includes basic strategy and advanced strategy. The basic strategy involves data input and output, standardization and basic mathematical logic operation. The initial data pipeline corresponds to the strategy one by one, and the data pipeline is defined by the strategy for data transmission and communication, and the specific formula is as follows:

[0013] U i ={C i ,S(C i ),O(C i ,S(C i ))}

[0014] Wherein, U i represents the ith initial data unit, C i represents the ith information core, and the key information extracted from the database. S(C i ) represents the strategy function based on the information core C i , O(C i , S(C i )) represents the data pipeline function, and the data pipeline is generated according to the information core and the strategy. I represents the index of the data unit. The data standardization strategy is specifically represented by the formula:

[0015]

[0016] Wherein, S norm (C i ) represents the data standardization strategy function applied to the information core C i , and norm represents the standardization operation. Represented as information core C i The mean, Represented as information core C i The standard deviation is then determined. Secondly, the data pipelines are randomly associated, and information kernels containing new information are continuously generated to construct higher-level data units. The initial data unit's data pipeline set is formed through all data pipelines of the initial data unit. Based on this, random association of data pipelines is performed to generate information kernels for binary combined data units. The specific formula is as follows:

[0017]

[0018] Where R represents the set of random association results from the data pipeline, r(O p O q ) represents the data pipeline O p and O q The association results, O p Represented as the p-th element in the data pipeline, O q Let p represent the q-th element in the data pipeline, where p and q represent the indices of the elements in the data pipeline. This is represented as a full-scale quantification operation. Then, by constructing a reward and punishment model and training the model by simulating human scoring of information cores, intelligent scoring and storage of the fused multimodal data are achieved. By automatically filtering out information cores with practical significance, high-quality data information is provided for the digital integrated application system based on intelligent highway management. The specific formula of the reward and punishment model is expressed as follows:

[0019]

[0020] Where Loss(Θ) represents the loss function of the reward-penalty model, Θ represents the parameter set of the reward-penalty model, x represents the data sample of the combined data unit, X represents the set of training samples, and r Θ (x) represents the prediction of the data sample x by the reward and punishment model based on the parameter set Θ, where ∈ represents the regularization parameter. Let x be the actual label of the data sample, σ be the sigmoid function, and log be the logarithmic function. Finally, a task-driven intelligent information extraction method is constructed. Using the actual task request as input, a topic model based on natural language understanding decomposes the task request into a set of topic titles for retrieving information kernels. Then, through random association in the data pipeline, a set of information kernels of combined data units related to the task request is generated. Through training the constructed reward-penalty model, intelligent ranking of the information kernels is achieved, storing the optimal matching information kernel for a given task request. The specific formula is expressed as follows:

[0021]

[0022] wherein, Loss Request (Θ) represents the loss function of the actual task demand Request, Request represents the actual task demand, represents the expected value operation, represents the expected operation on all possible task requests and information cores, C β represents the βth information core, C φ represents the φth information core, β and φ represent the index of the information core, D represents the set of training samples, each sample is a triple (Request, c β , c φ ), from the pairing of task requests and information cores, Rank Θ (Request, c β ) and Rank Θ (Request, c φ ) respectively represent the ranking of the information cores c β and c φ in the ranking list under the given task request Request, the model parameters Θ are optimized by minimizing the loss function Loss Request (Θ) so that the model can realize intelligent and autonomous extraction of the fused multi-modal data information.

[0023] Preferably, the digital integrated application module includes an intelligent analysis and application unit, which performs in-depth analysis on the processed data through the effective information of the multi-modal data extracted by the data fusion processing unit, ensures the generation of valuable application schemes, and provides intelligent support for highway management.

[0024] Preferably, the highway wisdom management and maintenance module includes a highway state monitoring unit, which monitors various state data of the highway in real time, including pavement temperature, humidity, crack width, settlement height, and traffic flow, to ensure that potential problems and risks can be discovered in a timely manner.

[0025] Preferably, the highway wisdom management and maintenance module includes a maintenance strategy optimization unit, which proposes a highway maintenance strategy optimization algorithm based on a genetic algorithm, analyzes historical data and current state indicators, predicts future highway maintenance needs, and ensures that reasonable maintenance strategies can be developed in advance to improve the resource utilization efficiency of the highway.

[0026] Specifically, the highway maintenance strategy optimization algorithm based on the genetic algorithm is as follows: first, a target function mathematical model is constructed to quantify the decision-making problem of highway maintenance, and a decision variable y irepresents whether the ith road is selected as the maintenance object, and the value of 1 indicates that the maintenance is selected, and the value of 0 indicates that the maintenance is not selected, and the objective function mathematical model is specifically expressed as:

[0027]

[0028] Wherein, F represents the objective function, aiming to minimize the total maintenance cost while maximizing the maintenance quality, min represents the minimization function, c i represents the maintenance cost of the ith road, represents the economic investment required for the maintenance of the ith road, i represents the index of the number of roads, N represents the total number of roads, s.t. represents the conditional function, Quality represents the total of the maintenance quality, used to reflect the overall quality effect of the maintenance strategy, max represents the maximization function, b i represents the measurement of the road maintenance quality, the actual maintenance demand is converted into a mathematical expression solved by an algorithm through the constructed objective function mathematical model, ensuring that the model can effectively reflect the economy and efficiency of the road maintenance strategy, further, considering that the maintenance cost will also be affected by time and space, the influence of time mainly includes the real-time fluctuation of labor cost and material cost, the influence of space mainly includes the cooperative maintenance demand of adjacent road sections, the time decay factor and the space correlation term are introduced into the original objective function F, and the updated objective function is F(t), which is expressed as follows:

[0029]

[0030] Wherein, t is time, c i (t) is time-varying maintenance cost, a is a recent maintenance penalty coefficient, which represents the economic penalty strength of repeated maintenance of the same road section in a short period, b is a time decay rate, which is used to control the decay rate of the penalty term with time interval (t-t last ), (t-t last ) is a time interval, f ij is a space coupling coefficient, E is a road network topology edge set, y i (t) represents whether the ith road is selected as the maintenance object at time t, y j (t) represents whether the jth road is selected as the maintenance object at time t, a compound Poisson-Wiener process is introduced to enhance the random disturbance term, which is expressed as follows:

[0031]

[0032] Wherein, k is a road surface natural recovery coefficient, reflecting the self-repairing ability of the road surface without traffic load, e is a traffic flow fluctuation intensity coefficient, quantifying the randomness of the accumulation of small damage of the road surface caused by normal traffic flow, W tP is the standard Brownian motion for Wiener process, to simulate the continuous high-frequency micro-impacts generated by randomly distributed light vehicles in traffic flow t P is the burst firing counting process, to represent the expected value of heavy vehicle passing times per unit time that can cause significant damage t ξ is subject to Poisson distribution k τ is the single impact damage intensity, subject to Pareto distribution k δ(t-τ) is the impact occurrence time k is represented as Dirac impulse function, then, to provide a starting point for the iterative search process of genetic algorithm, to generate a sufficient diversity of initial solution set, to ensure that genetic algorithm can cover a wide range of solution space, the specific formula is:

[0033] S(0) = y1(0), y2(0), …, y P (0) = {y i (0)|i = 1, 2 …, P}

[0034] Where S(0) represents the initial population, including the starting P solutions of genetic algorithm, y1(0), y2(0), …, y P (0) represents the independent solution in the P initial population, y i (0) represents the i-th solution in the initial population, i represents the index of population size, ranging from 1 to P, P represents the population size, representing the total number of solutions in the initial population, by initializing the population to generate a set of initial solutions based on heuristic rules, the initial solution constitutes the first generation of the population of genetic algorithm, the diversity of the population directly affects the search ability of the algorithm and the quality of the final solution, the construction of a good initial population helps the algorithm to effectively explore the solution space and avoid falling into local optimal solution, secondly, the fitness of each solution is quantitatively evaluated to determine the pros and cons of the solution, and the solution with greater contribution to the optimization target is selected as the candidate for subsequent iteration, the specific formula is:

[0035] f(y) = w1·Cost(y) + w2·QUL(y)

[0036] wherein f(y) represents a fitness function for evaluating the overall performance of a solution y, y represents an individual solution in the population, w1 represents a cost weight coefficient for adjusting the degree of influence of the maintenance cost in the fitness function, Cost(y) represents a maintenance cost function representing the total maintenance cost calculated according to the solution y, w2 represents a quality weight coefficient for adjusting the degree of influence of the maintenance quality in the fitness function, and QUL(y) represents a maintenance quality function representing the maintenance quality calculated according to the solution y. Through the fitness function f(y), the genetic algorithm can find a balance point so that the total cost is minimized while the maintenance quality reaches a standard-compliant level. Then, through a selection mechanism, excellent individual solutions are selected from the current population according to the fitness, and the excellent individual solutions will be used as the parents of the next generation of the genetic algorithm. The specific formula is as follows:

[0037]

[0038] wherein Probability(y j ) represents the probability of selecting the individual solution y j , which represents a relative measure based on the fitness of the individual, and is used in the roulette wheel selection mechanism, y j represents the jth individual solution in the current population, j represents the index of the individual solution in the current population, and f(y j ) represents the fitness function of the individual y j . Through the selection principle based on fitness, the individual solution with a higher fitness has a high probability of being selected, so that the algorithm can select excellent solutions in the iteration process while maintaining the diversity of the overall algorithm. Finally, new genetic characteristics are introduced into the population through crossover and mutation operations to increase the diversity of the population and improve the probability of finding the global optimal solution. The specific formula is as follows:

[0039] Offspring=Crossover(y a ,y b )

[0040] y Offspring,变异a =Mutate(y a )

[0041] y Offspring,变异b =Mutate(y b )

[0042] wherein Offspring represents a newly generated offspring individual representing the result of the crossover operation, Crossover(y a ,y b ) represents a crossover operation for generating offspring individuals from two parent individual solutions y a and y bExchange genetic information, produce new offspring, y a With y b Indicated as the parent individual solution, a and b indicate the parent individual solution index, y Offspring,变异1 With y Offspring,变异2 Indicated as the offspring individual solution after mutation operation, Mutate(y a ) and Mutate(y b ) are indicated as mutation operation functions, new genetic characteristics are introduced through mutation operation, premature convergence of individual solutions is prevented, crossover operation creates offspring by combining the characteristics of two parents, mutation operation introduces new genetic diversity by randomly changing the characteristics of offspring, and the algorithm can better search the solution space by combining crossover operation and mutation operation to find the optimal solution of the problem.

[0043] Preferably, the data storage and management module stores and manages the collected multi-modal data by centralization, ensures the security and availability of the data, supports the storage and retrieval of big data, provides query and analysis functions of historical data, and provides data support for long-term management and planning of highways.

[0044] Preferably, the user monitoring and decision-making module provides an intuitive user interface and decision support functions, including data visualization tools and report generation functions, helps management personnel make correct decisions, and ensures that highway management personnel can conveniently use the system to query data, view analysis results and receive early warning information.

[0045] Compared with the prior art, the beneficial effects of the present application are:

[0046] 1、Data fusion processing unit proposes a highway information extraction algorithm based on multi-modal data fusion. This algorithm realizes the fusion and intelligent extraction of multi-modal data by constructing data units containing information cores, strategy sets, and data pipeline sets. The organization structure of each data unit includes an information core carrying key information, a strategy set including input-output strategies, data standardization strategies, and mathematical logic operation strategies, and a data pipeline set for information transmission and communication, thereby ensuring efficient processing and utilization of data. First, the initial data unit is directly associated with the database, and the initial information core is constructed by extracting the field name and technical name in the database, and combined with basic and advanced strategies, the data pipeline is defined to ensure the consistency and integrity of the data. Based on these initial data units, the system can effectively standardize and process data, making data comparable and operable between different sources. Through data standardization strategies, the system can eliminate the bias between different data sources, ensuring the accuracy and reliability of the data. Next, through the random association of data pipelines, the system generates high-level data unit information cores, realizing the deep fusion of multi-source data. This method not only improves the utilization efficiency of data, but also enables the system to extract valuable information cores from a large amount of data, thereby supporting more intelligent decision-making and prediction. The randomly associated data pipeline can construct multi-element combined data units, further enriching the system's data hierarchy structure and providing strong data support for intelligent highway management. In addition, based on the task-driven intelligent information extraction method, the actual task request is decomposed into a set of topic titles for retrieving information cores through a topic model of natural language understanding, and a set of combined data unit information cores related to the task request is generated through the random association of data pipelines. By constructing a reward and punishment model and simulating manual scoring methods, the system can realize intelligent scoring and sorting of the fused multi-modal data, ensuring that the extracted information cores have practical significance and high quality. The intelligent information extraction method can improve the response speed of the system and significantly enhance the flexibility and accuracy of the system in dealing with complex task requests. Through the target function, the economic decay effect in the time dimension and the spatial correlation constraint are included in the cost calculation, so that the maintenance decision not only considers the current input-output ratio, but also intelligently avoids repeated investment waste on the same road section in the short term. By introducing alpha and beta parameters, the system can automatically identify economically inefficient behaviors in historical maintenance records. The introduction of the compound Poisson-Wiener process accurately depicts the differential damage mechanism of traffic load on the road, where the continuous Brownian motion term captures the uniform wear caused by daily traffic, and the Poisson jump process simulates sudden impact events such as overloaded vehicles. This two-level stochastic modeling realizes the differentiation and early warning of "chronic damage" and "acute trauma" in maintenance strategies.In summary, the highway information extraction algorithm based on multi-modal data fusion provides strong technical support for highway intelligent management and maintenance by constructing data units, implementing data fusion and intelligent extraction. This algorithm significantly improves the intelligent level of the system while ensuring efficient processing and utilization of data, meeting the demand for high-quality data and intelligent decision-making in highway management and maintenance. Through the highway information extraction algorithm based on multi-modal data fusion, the highway intelligent management and maintenance system can better achieve real-time monitoring, predictive maintenance, and resource optimization, thereby improving the management efficiency and maintenance effect of highways.

[0047] 2. The maintenance strategy optimization unit proposes a highway maintenance strategy optimization algorithm based on genetic algorithm. First, the algorithm quantifies the decision-making problem of highway maintenance by constructing a mathematical model of the objective function, ensuring that the model effectively reflects the economy and efficiency of the maintenance strategy. The objective function aims to minimize the total maintenance cost while maximizing the maintenance quality. By converting the actual maintenance needs into a mathematical expression for algorithm solving, the decision-making is more scientific and reasonable. Second, the initial population generation step of the genetic algorithm ensures extensive coverage of the solution space. By generating a sufficient diversity of initial solution sets, the algorithm can explore different maintenance strategies from multiple angles, avoiding local optimal solutions and improving the global search ability and quality of the final solution. Third, the fitness evaluation mechanism quantifies the performance of each solution, selecting solutions that contribute more to the optimization target. The fitness function considers both maintenance cost and maintenance quality, allowing the algorithm to balance these two key factors during iteration, ensuring that costs are controlled while maintaining quality meets standards. In terms of selection mechanism, the selection principle based on fitness gives high fitness individuals a higher probability of being selected. This mechanism not only selects excellent solutions but also maintains population diversity, preventing premature convergence and improving the probability of finding global optimal solutions. Finally, through crossover and mutation operations, new genetic characteristics are introduced, further increasing population diversity. Crossover operation creates offspring by combining the characteristics of two parent individuals, while mutation operation introduces new genetic diversity by randomly changing the characteristics of offspring. The combination of these two operations allows the algorithm to better explore the solution space and find the optimal maintenance strategy. In summary, the highway maintenance strategy optimization algorithm based on genetic algorithm provides efficient and intelligent decision support for highway maintenance in a digital integrated application system based on highway intelligent management and maintenance. This algorithm not only improves the scientificity and rationality of decision-making, reducing maintenance costs, but also enhances maintenance quality, ultimately achieving intelligent management of highway maintenance. BRIEF DESCRIPTION OF DRAWINGS

[0048] The invention is further illustrated by the accompanying drawings, but the embodiments in the drawings do not constitute any limitation to the invention, and other embodiments can be obtained by those skilled in the art without creative effort on the basis of the following drawings.

[0049] Figure 1 The structural schematic diagram of the present invention. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to 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 the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those skilled in the art without creative effort belong to the scope of protection of the present invention.

[0051] Referring to Figure 1 The present invention provides a digital integrated application system based on intelligent road management and maintenance, including data acquisition and sensing module, data transmission and communication module, digital integrated application module, intelligent road management and maintenance module, data storage and management module, user monitoring and decision-making module, characterized in that: the data acquisition and sensing module is used to collect real-time road state data through a sensor network, including temperature, humidity, pressure and vibration information, the data transmission and communication module is used to transmit the collected data through wireless network and 5G technology and communicate with the central server, the digital integrated application module includes a data fusion processing unit and an intelligent analysis and application unit, the data fusion processing unit proposes a multi-modal data fusion-based road information extraction algorithm for data fusion processing and intelligent extraction of effective information from multi-modal data from the sensor network, the intelligent analysis and application unit is used to intelligently analyze the effective information extracted from the fused data and apply the analysis results to road state prediction, maintenance strategy optimization and decision support, the intelligent road management and maintenance module includes a road state monitoring unit and a maintenance strategy optimization unit, the road state monitoring unit is used to monitor real-time road state indicators, including road surface temperature, humidity, crack width, settlement height and traffic flow, the maintenance strategy optimization unit proposes a genetic algorithm-based road maintenance strategy optimization algorithm for developing and optimizing road maintenance strategies to improve maintenance effectiveness and resource utilization efficiency, the data storage and management module is used to centrally store and manage collected data and processed analysis data to ensure data security and availability, the user monitoring and decision-making module is used to provide users with real-time monitoring interfaces and decision support functions, including data visualization and early warning notification and decision control functions.

[0052] Referring to Figure 1Further, the data acquisition and sensing module collects multi-modal data of the highway in real time through the sensor network constructed on the highway, including temperature sensors, humidity sensors, pressure sensors, and vibration sensors, ensuring that the state information of the highway can be obtained in real time, providing a basis for subsequent data processing and analysis.

[0053] Referring to Figure 1 Further, the data transmission and communication module transmits the collected data to the central server in a timely manner through the use of wireless networks and 5G communication technology, ensuring the real-time and accuracy of the data to support immediate decision-making and reaction.

[0054] Referring to Figure 1 Further, the digital integration application module includes a data fusion processing unit that proposes a highway information extraction algorithm based on multi-modal data fusion, which integrates and processes multi-modal data from different sensors to ensure data consistency and integrity, and extracts effective information and value from multi-modal data.

[0055] Referring to Figure 1 Further, the highway information extraction algorithm based on multi-modal data fusion is as follows: first, assume a data unit organization structure, which is the basic unit of the overall multi-modal data fusion algorithm, each data unit contains an information core, a strategy set, and a data pipeline set corresponding to the strategy, and the data unit organization structure is specifically represented by the formula:

[0056] U = {C, S, O}

[0057] Where U represents the data unit, C represents the information core, S represents the strategy set, and O represents the data pipeline set. The information core C is the core component of the data unit, carrying the key information for data intelligence implementation. The strategy set S includes input-output strategies, data standardization strategies, and mathematical logic operation strategies. The data pipeline O is the channel for information transmission and communication, corresponding to the strategy one by one, and generates a new information core through the association of the data pipeline, realizing the hierarchical structure of the data unit. Then, the initial data unit is directly associated with the database, including the construction of the initial information core, the construction of the initial strategy, and the construction of the initial data pipeline. The initial information core is the field name and technical name extracted from the database, which is the common attribute of the data. The initial strategy set includes basic strategies and advanced strategies. The basic strategies involve data input-output, standardization, and basic mathematical logic operations. The initial data pipeline corresponds to the strategy one by one, and defines the data pipeline for data transmission and communication through the strategy. The specific formula is:

[0058] U i = {C i ,S(C i), O(C i ,S(C i ))

[0059] wherein, U i represents the i-th initial data unit, C i represents the i-th information core, the key information extracted from the database, S(C i ) represents the strategy function based on the information core C i , O(C i ,S(C i )) represents the data pipeline function, and the data pipeline is generated according to the information core and the strategy, i represents the index of the data unit, and the data standardization strategy is specifically represented by the formula:

[0060]

[0061] wherein, S norm (C i ) represents the data standardization strategy function applied to the information core C i , norm represents the standardization operation, represents the mean value of the information core C i , represents the standard deviation of the information core C i , secondly, the data pipeline is randomly associated, and the information core of the high-level data unit containing new information is continuously generated to construct the high-level data unit, and the data pipeline set of the initial data unit is formed through all the data pipelines of the initial data unit, and the random association of the data pipeline is carried out on this basis to generate the information core of the binary combined data unit, and the specific formula is represented as:

[0062]

[0063] wherein, R represents the random association result set of the data pipeline, r(O p ,O q ) represents the association result of the data pipeline O p and O q , O p represents the p-th element in the data pipeline, and O q represents the q-th element in the data pipeline, and p and q represent the indices of the elements in the data pipeline, represents the universal quantization operation, then, the reward and punishment model is constructed, and the model is trained by simulating the artificial scoring method of the information core, realizing the intelligent scoring and storage of the fused multi-modal data, and the information core with practical significance is automatically screened out, providing high-quality data information for the digital integrated application system based on intelligent management of highways, and the reward and punishment model is specifically represented by the formula:

[0064]

[0065] wherein, Loss(Θ) represents the loss function of the reward-punishment model, Θ represents the parameter set of the reward-punishment model, x represents the data sample of the combined data unit, X represents the set of training samples, r Θ (x) represents the prediction of the data sample x by the reward-punishment model according to the parameter set Θ, ∈ represents the regularization parameter, represents the actual label of the data sample x, σ represents the Sigmoid function, log represents the logarithmic function, finally, a task-driven intelligent information extraction method is constructed, taking the actual task request as the information input, using the topic model of natural language understanding to decompose the task request into a set of topic titles of the information core, then generating a set of information cores of the combined data unit related to the task request through random association of the data pipeline, through the training of the constructed reward-punishment model, the intelligent sorting of the information core is realized, and the optimal matching information core is stored for the given task request, and the specific formula is represented as:

[0066]

[0067] wherein, Loss Request (Θ) represents the loss function of the reward-punishment model in the actual task demand Request, Request represents the actual task demand, represents the expected value operation, represents the expected operation on all possible task requests and information core pairs, C β represents the βth information core, C φ represents the φth information core, β and φ represent the index of the information core, D represents the set of training samples, each sample is a three-tuple (Request, c β ,c φ ), from the task request and the information core pair, Rank Θ (Request,c β ) and Rank Θ (Request,c φ ) respectively represent the ranking of the information core c β and c φ in the ranking list under the given task request Request, by minimizing the loss function Loss Request (Θ) to optimize the model parameter Θ, so that the model can realize the intelligent autonomous extraction of the fused multi-modal data information.

[0068] Referring to Figure 1Further, the digital integrated application module includes an intelligent analysis and application unit that performs in-depth analysis on the processed data based on the multi-modal data effective information extracted by the data fusion processing unit, ensuring the generation of valuable application schemes and providing intelligent support for highway management.

[0069] Referring to Figure 1 Further, the highway intelligent management and maintenance module includes a highway state monitoring unit that monitors various state data of the highway in real time, including pavement temperature, humidity, crack width, settlement height, and traffic flow, ensuring that potential problems and risks can be discovered in a timely manner.

[0070] Referring to Figure 1 Further, the highway intelligent management and maintenance module includes a maintenance strategy optimization unit that proposes a genetic algorithm-based highway maintenance strategy optimization algorithm, which analyzes historical data and current state indicators to predict future highway maintenance needs, ensuring that reasonable maintenance strategies can be developed in advance and improving the resource utilization efficiency of highways.

[0071] Referring to Figure 1 Further, the genetic algorithm-based highway maintenance strategy optimization algorithm is as follows: first, a target function mathematical model is constructed to quantify the decision-making problem of highway maintenance, and decision variables y i are defined, where y i represents whether the i-th highway is selected as a maintenance object, and takes a value of 1 when maintenance is selected and a value of 0 when maintenance is not selected, and the specific formula of the target function mathematical model is:

[0072]

[0073] where F represents the target function, aiming to minimize the total maintenance cost while maximizing the maintenance quality, min represents the minimization function, c i represents the maintenance cost of the i-th highway, represents the economic input required for the maintenance of the i-th highway, i represents the highway number index, N represents the total number of highways, s.t. represents the conditional function, Quality represents the total sum of maintenance quality, reflecting the overall quality effect of the maintenance strategy, max represents the maximization function, b iThe highway maintenance quality is expressed as a measure, the actual maintenance demand is converted into a mathematical expression solved by an algorithm through the constructed objective function mathematical model, ensuring that the model can effectively reflect the economy and efficiency of the highway maintenance strategy. Furthermore, considering that the maintenance cost is also affected by time and space, the time impact mainly includes the real-time fluctuations of labor costs and material costs, and the space impact mainly includes the coordinated maintenance demand of adjacent road sections. The time decay factor and space correlation term are introduced into the original objective function F, and the updated objective function is F(t), which is expressed as follows:

[0074]

[0075] Where t is time, c i (t) is the time-varying maintenance cost, α is the recent maintenance penalty coefficient, which represents the economic penalty strength of repeated maintenance of the same road section in a short period of time, β is the time decay rate, which controls the decay rate of the penalty term with time interval (t-t last ), (t-t last ) is the time interval, φ ij is the space coupling coefficient, E is the road network topology edge set, y i (t) indicates whether the ith highway is selected as the maintenance object at time t, y j (t) indicates whether the jth highway is selected as the maintenance object at time t, the compound Poisson-Wiener process is introduced to enhance the random disturbance term, which is expressed as follows:

[0076]

[0077] Where κ is the pavement natural recovery coefficient, reflecting the self-repairing ability of the pavement under no traffic load, ε is the traffic flow fluctuation intensity coefficient, quantifying the randomness of the accumulation of small damage to the pavement caused by normal traffic flow, W t is the standard Brown motion of Wiener process, simulating the continuous high-frequency small impact generated by randomly distributed light vehicles in traffic flow, P t is the burst impact counting process, used to represent the expected number of times of heavy vehicles passing through per unit time that cause significant damage, P t obeys Poisson distribution, ξ k is the single impact damage intensity, which obeys Pareto distribution, τ k is the impact occurrence time, δ(t-τ k ) is the Dirac impulse function, then, a starting point is provided for the iterative search process of genetic algorithm to generate a sufficient diversity of initial solution set, ensuring that the genetic algorithm can cover a wide range of solution space, and the specific formula is expressed as:

[0078] S(0) = y1(0), y2(0), …, y P(0)={y i (0)|i=1,2…,P}

[0079] Where S(0) represents the initial population, containing the P solutions of the genetic algorithm, y1(0), y2(0), ..., y P (0) represents the independent solutions in the P initial populations, y i (0) represents the i-th solution in the initial population, where i represents the index of the population size, ranging from 1 to P, and P represents the population size, representing the total number of solutions in the initial population. By initializing the population, a set of initial solutions based on heuristic rules is generated. The initial solutions constitute the first generation population of the genetic algorithm. The diversity of the population directly affects the search ability of the algorithm and the quality of the final solution. A well-constructed initial population helps the algorithm to effectively explore the solution space and avoid getting trapped in local optima. Secondly, the fitness of each solution is quantitatively evaluated to determine the quality of the solution and select solutions that contribute more to the optimization objective as candidates for subsequent iterations. The specific formula is expressed as:

[0080] f(y) = w1·Cost(y) + w2·QUL(y)

[0081] Where f(y) represents the fitness function, used to evaluate the overall performance of solution y, y represents an independent solution in the population, w1 represents the cost weight coefficient, used to adjust the influence of maintenance cost in the fitness function, Cost(y) represents the maintenance cost function, representing the total maintenance cost calculated based on solution y, w2 represents the quality weight coefficient, used to adjust the influence of maintenance quality in the fitness function, and QUL(y) represents the maintenance quality function, representing the maintenance quality calculated based on solution y. Through the fitness function f(y), the genetic algorithm can find an equilibrium point that minimizes the total cost while also achieving a standard maintenance quality. Then, through a selection mechanism, excellent individual solutions are selected from the current population based on fitness. These excellent individual solutions will serve as the parents of the next generation of the genetic algorithm. The specific formula is expressed as follows:

[0082]

[0083] Among them, Probability(y j ) represents the individual solution y j The probability of being selected represents a relative measure based on individual fitness, used in the roulette wheel selection mechanism, y j Let f(y) represent the j-th individual solution in the current population, where j represents the index of the individual solution in the current population. j ) represents an individual y jThe fitness function of the adaptive function is selected by the principle of fitness-based selection. The higher the probability of selecting the individual solution, the higher the probability of selecting the individual solution. The algorithm can select excellent solutions in the iteration process, while maintaining the diversity of the overall algorithm. Finally, new genetic characteristics are introduced in the population through crossover and mutation operations to increase the diversity of the population and improve the probability of finding the global optimal solution. The specific formula is:

[0084] Offspring=Crossover(y a ,y b )

[0085] y Offspring,变异a =Mutate(y a )

[0086] y Offspring,变异b =Mutate(y b )

[0087] where Offspring represents the newly generated offspring, Crossover(y a ,y b ) represents the crossover operation, which exchanges genetic information from two parent individual solutions y a and y b to produce new offspring, y a and y b represent the parent individual solutions, a and b represent the parent individual solution index, y Offspring,变异1 and y Offspring,变异2 represent the offspring individual solution after mutation, Mutate(y a ) and Mutate(y b ) represent the mutation operation function, which introduces new genetic characteristics through mutation operation to prevent premature convergence of individual solutions. The crossover operation creates offspring by combining the characteristics of two parents, and the mutation operation introduces new genetic diversity by randomly changing the characteristics of the offspring. By combining crossover and mutation operations, the algorithm can better search the solution space and find the optimal solution to the problem.

[0088] Referring to Figure 1 , further, the data storage and management module stores and manages the collected multi-modal data, ensuring the security and availability of the data, while supporting the storage and retrieval of big data, providing query and analysis functions for historical data, and providing data support for long-term management and planning of highways.

[0089] Referring to Figure 1Further, the user monitoring and decision-making module helps managers make correct decisions by providing an intuitive user interface and decision support functions, including data visualization tools and report generation functions, while ensuring that highway managers can easily use the system for data query, analysis result viewing and early warning information reception.

[0090] In specific use, first, the data acquisition and sensing module is used to collect real-time state data of the highway through the sensor network, including temperature, humidity, pressure and vibration information, the data transmission and communication module is used to transmit the collected data through wireless network and 5G technology and communicate with the central server, then, the digital integrated application module includes a data fusion processing unit and an intelligent analysis and application unit, the data fusion processing unit proposes a highway information extraction algorithm based on multi-modal data fusion for data fusion processing of multi-modal data from the sensor network and intelligent extraction of effective information, the intelligent analysis and application unit is used for intelligent analysis of the effective information extracted from the fused data and application of the analysis results to highway state prediction, maintenance strategy optimization and decision support, secondly, the intelligent highway management and maintenance module includes a highway state monitoring unit and a maintenance strategy optimization unit, the highway state monitoring unit is used to monitor specific state indicators of the highway in real time, including pavement temperature, humidity, crack width, settlement height and traffic flow, the maintenance strategy optimization unit proposes a highway maintenance strategy optimization algorithm based on genetic algorithm for formulating and optimizing the maintenance strategy of the highway to improve the maintenance effect and resource utilization efficiency, finally, the data storage and management module is used to centrally store and manage the collected data and processed analysis data, ensuring the safety and availability of the data, and the user monitoring and decision-making module is used to provide real-time monitoring interface and decision support functions for users, including data visualization and early warning notification and decision control functions.

[0091] Compared with the prior art, the beneficial effects of the present application are:

[0092] 3、Data fusion processing unit proposes a highway information extraction algorithm based on multi-modal data fusion. This algorithm realizes the fusion and intelligent extraction of multi-modal data by constructing data units containing information cores, strategy sets, and data pipeline sets. The organizational structure of each data unit includes an information core carrying key information, a strategy set including input-output strategies, data standardization strategies, and mathematical logic operation strategies, and a data pipeline set for information transmission and communication, thereby ensuring efficient processing and utilization of data. First, the initial data unit is directly associated with the database, and the initial information core is constructed by extracting the field name and technical name in the database, and combined with basic and advanced strategies, the data pipeline is defined to ensure the consistency and integrity of the data. Based on these initial data units, the system can effectively standardize and process data, making data comparable and operable between different sources. Through data standardization strategies, the system can eliminate the bias between different data sources, ensuring the accuracy and reliability of the data. Next, through the random association of data pipelines, the system generates high-level data unit information cores, realizing the deep fusion of multi-source data. This method not only improves the utilization efficiency of data, but also enables the system to extract valuable information cores from a large amount of data, thereby supporting more intelligent decision-making and prediction. The randomly associated data pipeline can construct multi-element combined data units, further enriching the system's data hierarchy and providing strong data support for intelligent highway management. In addition, based on the task-driven intelligent information extraction method, the actual task request is decomposed into a set of topic titles for retrieving information cores through a topic model of natural language understanding, and a set of combined data unit information cores related to the task request is generated through the random association of data pipelines. By constructing a reward and punishment model and simulating manual scoring methods, the system can realize intelligent scoring and sorting of the fused multi-modal data, ensuring that the extracted information cores have practical significance and high quality. The intelligent information extraction method can improve the response speed of the system and significantly enhance the flexibility and accuracy of the system in dealing with complex task requests. Through the target function, the economic decay effect in the time dimension and the spatial correlation constraint are incorporated into the cost calculation, so that the maintenance decision not only considers the current input-output ratio, but also intelligently avoids repeated investment waste on the same road segment in the short term. By introducing alpha and beta parameters, the system can automatically identify economically inefficient behaviors in historical maintenance records. The introduction of the compound Poisson-Wien process accurately depicts the differential damage mechanism of traffic load on the road, where the continuous Brownian motion term captures the uniform wear caused by daily traffic, and the Poisson jump process simulates sudden impact events such as overloaded vehicles. This two-level stochastic modeling realizes the differentiation and early warning of "chronic damage" and "acute trauma" in maintenance strategies.In summary, the highway information extraction algorithm based on multi-modal data fusion provides strong technical support for highway intelligent management and maintenance by constructing data units, implementing data fusion and intelligent extraction. This algorithm significantly improves the intelligent level of the system while ensuring efficient processing and utilization of data, meeting the demand for high-quality data and intelligent decision-making in highway management and maintenance. Through the highway information extraction algorithm based on multi-modal data fusion, the highway intelligent management and maintenance system can better achieve real-time monitoring, predictive maintenance, and resource optimization, thereby improving the management efficiency and maintenance effect of highways.

[0093] 4. The maintenance strategy optimization unit proposes a highway maintenance strategy optimization algorithm based on genetic algorithm. First, the algorithm quantifies the decision-making problem of highway maintenance by constructing a mathematical model of the objective function, ensuring that the model can effectively reflect the economy and efficiency of the maintenance strategy. The objective function aims to minimize the total maintenance cost while maximizing the maintenance quality. By converting the actual maintenance demand into a mathematical expression for algorithm solving, the decision-making is more scientific and reasonable. Second, the initial population generation step of the genetic algorithm ensures extensive coverage of the solution space. By generating a sufficient diversity of initial solution sets, the algorithm can explore different maintenance strategies from multiple angles, avoiding local optimal solutions and improving the global search ability of the algorithm and the quality of the final solution. Third, the fitness evaluation mechanism quantifies the performance of each solution and selects solutions that contribute more to the optimization target. The fitness function considers both maintenance cost and maintenance quality, allowing the algorithm to balance these two key factors during the iteration process, ensuring that costs are controlled while maintaining quality meets standards. In terms of selection mechanism, the selection principle based on fitness gives high fitness individuals a higher probability of being selected. This mechanism not only selects excellent solutions but also maintains the diversity of the population, preventing premature convergence of the algorithm and improving the probability of discovering the global optimal solution. Finally, through crossover and mutation operations, new genetic characteristics are introduced, further increasing the diversity of the population. Crossover operation creates offspring by combining the characteristics of two parent individuals, while mutation operation introduces new genetic diversity by randomly changing the characteristics of offspring. The combination of these two operations allows the algorithm to better explore the solution space and find the optimal maintenance strategy. In summary, the highway maintenance strategy optimization algorithm based on genetic algorithm provides efficient and intelligent decision support for highway maintenance in a digital integrated application system based on highway intelligent management and maintenance. This algorithm not only improves the scientificity and rationality of decision-making and reduces maintenance costs but also enhances maintenance quality, ultimately achieving intelligent management of highway maintenance.

[0094] Although the present application has been described in detail with reference to the foregoing embodiments, the technical solutions recorded in the foregoing embodiments can be modified, or some of the technical features can be replaced by equivalent features, by those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A digital integrated application system based on intelligent highway maintenance, comprising a data acquisition and sensing module, a data transmission and communication module, a digital integrated application module, an intelligent highway maintenance module, a data storage and management module, and a user monitoring and decision-making module, characterized in that: The data acquisition and sensing module is used to collect real-time highway status data, including temperature, humidity, pressure, and vibration information, through a sensor network. The data transmission and communication module is used to transmit the collected data and communicate with the central server via wireless network and 5G technology. The digital integration application module includes a data fusion processing unit and an intelligent analysis and application unit. The data fusion processing unit proposes a highway information extraction algorithm based on multimodal data fusion to perform data fusion processing on multimodal data from the sensor network and intelligently extract effective information. The intelligent analysis and application unit is used to intelligently analyze the effective information extracted from the fused data and apply the analysis results to highway status prediction, maintenance strategy optimization, and decision support. The smart highway management module includes a highway status monitoring unit and a maintenance strategy optimization unit. The highway status monitoring unit is used to monitor specific highway status indicators in real time, including pavement temperature, humidity, crack width, settlement height, and traffic flow. The maintenance strategy optimization unit proposes a highway information extraction algorithm based on a genetic algorithm. The maintenance strategy optimization algorithm is used to formulate and optimize highway maintenance strategies to improve maintenance effectiveness and resource utilization efficiency. The data storage and management module centrally stores and manages collected and processed analysis data, ensuring data security and availability. The user monitoring and decision-making module provides users with a real-time monitoring interface and decision support functions, including data visualization, early warning notifications, and decision control functions. The intelligent highway management module includes a highway condition monitoring unit and a maintenance strategy optimization unit. The highway condition monitoring unit monitors various highway condition data in real time, including pavement temperature, humidity, crack width, settlement height, and traffic flow, ensuring timely detection of potential problems and risks. The maintenance strategy optimization unit proposes a highway maintenance strategy optimization algorithm based on a genetic algorithm. By analyzing historical data and current condition indicators, it predicts future highway maintenance needs, ensuring the ability to formulate reasonable maintenance strategies in advance and improve highway resource utilization efficiency. First, a mathematical model of the objective function is constructed to quantify the decision problem of highway maintenance, defining decision variables. Represented as the first Whether a road is selected as a maintenance target is determined by a value of 1 (selection for maintenance) and 0 (no maintenance). The specific formula for the objective function mathematical model is as follows: in, Expressed as an objective function, it aims to minimize the total maintenance cost while maximizing the maintenance quality. Represented as a minimization function, Represented as the first The maintenance cost of the first highway represents the cost of maintaining the second highway. The economic investment required for the maintenance of this highway. Represented as a highway number index, Expressed as the total number of highways, Represented as a conditional function, It is expressed as the sum of maintenance quality, used to reflect the overall quality and effectiveness of maintenance strategies. Represented as a maximization function, This is expressed as a measure of highway maintenance quality. A mathematical model of the objective function transforms actual maintenance needs into mathematical expressions solvable by an algorithm, ensuring the model effectively reflects the economy and efficiency of highway maintenance strategies. Furthermore, considering that maintenance costs are also affected by time and space—including real-time fluctuations in labor and material costs, and space-related factors such as the collaborative maintenance needs of adjacent road sections—the original objective function... Introducing a time decay factor and a spatial correlation term, the updated objective function is: The expression is as follows: Where t is time, For time-varying maintenance costs, This is the recent maintenance penalty coefficient, used to characterize the intensity of the economic penalty for repeated maintenance on the same road section within a short period of time. The time decay rate is used to control the penalty term as the time interval increases. The decay rate, For time intervals, The spatial coupling coefficient is... For the road network topology edge set, Represented as the time t, the first... Whether this highway will be selected as a maintenance target Represented as the time t, the first... Whether a highway is selected as a maintenance target can be determined by introducing a composite Poisson-Wiener process and enhancing the random disturbance term, as follows: in, The road surface natural recovery coefficient reflects the road surface's ability to self-repair when there is no traffic load. This is the traffic flow fluctuation intensity coefficient, used to quantify the randomness of the accumulation of minor road surface damage caused by normal traffic flow. The Wiener process is a standard Brownian motion used to simulate continuous high-frequency minute impacts generated by randomly distributed light vehicles in traffic flow. This is a sudden impact counting process, used to represent the expected number of heavy vehicles passing through per unit time that cause significant damage. Follows a Poisson distribution. The single impact damage intensity follows a Pareto distribution. The moment of impact, The overall expression is represented by the Dirac impulse function, which then provides a starting point for the iterative search process of the genetic algorithm, generating a sufficiently diverse initial solution set to ensure that the genetic algorithm can cover a wide region of the solution space. The specific formula is expressed as follows: in, This is represented as the initial population, containing the starting data for the genetic algorithm. One solution. Represented as Independent solutions in the initial population Represented as the first in the initial population One solution. Indices representing population size, ranging from 1 to , The population size is represented by the total number of solutions in the initial population. Initializing the population generates a set of initial solutions based on heuristic rules. These initial solutions constitute the first generation of the genetic algorithm. The diversity of the population directly affects the algorithm's search ability and the quality of the final solution. A well-constructed initial population helps the algorithm effectively explore the solution space and avoid getting trapped in local optima. Next, the fitness of each solution is quantitatively evaluated to determine its quality, and solutions that contribute significantly to the optimization objective are selected as candidates for subsequent iterations. The specific formula is as follows: in, Represented as a fitness function, used to evaluate solutions. Overall performance This is represented as an independent solution in the population. This is represented as a cost weighting coefficient, used to adjust the degree of influence of maintenance costs in the fitness function. Represented as a maintenance cost function, it represents the cost based on the solution. Calculated total maintenance cost, This is represented as a quality weighting coefficient, used to adjust the degree of influence of maintenance quality in the fitness function. Represented as a maintenance quality function, it represents the result based on the solution. The calculated maintenance quality is obtained through the fitness function. Genetic algorithms can find an equilibrium point that minimizes total cost while ensuring maintenance quality meets standards. Then, through a selection mechanism, excellent individuals are selected from the current population based on fitness. These excellent individuals will serve as the parents of the next generation of the genetic algorithm. The specific formula is as follows: in, Represented as individual solutions The probability of being selected represents a relative measure based on an individual's fitness, used in the roulette wheel selection mechanism. Represented as the th in the current population Individual solutions This represents the index of an individual solution in the current population. Represented as an individual The fitness function, based on the fitness selection principle, ensures that solutions with higher fitness are selected with a higher probability. This allows the algorithm to select excellent solutions during iteration while maintaining the diversity of the overall algorithm. Finally, crossover and mutation operations introduce new genetic features into the population, increasing population diversity and improving the probability of finding the global optimum. The specific formula is as follows: in, The newly generated offspring individuals represent the result of the crossover operation. This is represented as a crossover operation, used to solve problems from two parent individuals. and They exchange genetic information and produce new offspring. and This is represented as the parent generation's individual solution. and This represents the parent individual's solution index. and This represents the solution of the offspring individual after the mutation operation. and The mutation operation function is used to introduce new genetic features to prevent premature convergence of individual solutions. The crossover operation creates offspring by combining the features of two parents, and the mutation operation introduces new genetic diversity by randomly changing the features of offspring. By combining the crossover and mutation operations, the algorithm can better search the solution space and find the optimal solution to the problem.

2. The digital integrated application system based on intelligent highway management and maintenance according to claim 1, characterized in that: The data acquisition and sensing module collects multimodal data of the highway in real time through a sensor network built on the highway, including temperature sensors, humidity sensors, pressure sensors and vibration sensors, to ensure that the highway status information can be obtained in real time, providing a foundation for subsequent data processing and analysis.

3. The digital integrated application system based on intelligent highway management and maintenance according to claim 1, characterized in that: The data transmission and communication module utilizes wireless network and 5G communication technology to transmit the collected data to the central server in a timely manner, ensuring the real-time nature and accuracy of the data to support immediate decision-making and response.

4. The digital integrated application system based on intelligent highway management and maintenance according to claim 1, characterized in that, The digital integrated application module includes a data fusion processing unit and an intelligent analysis and application unit. The data fusion processing unit proposes a highway information extraction algorithm based on multimodal data fusion. By integrating and processing multimodal data from different sensors, it ensures the consistency and integrity of the data and extracts the effective information and utilization value of the multimodal data. The intelligent analysis and application unit uses the effective information of multimodal data extracted by the data fusion processing unit to conduct in-depth analysis of the processed data, ensuring the generation of valuable application solutions and providing intelligent support for highway management.

5. A digital integrated application system based on intelligent highway management and maintenance according to claim 4, characterized in that: First, assuming the organizational structure of a data unit, the data unit constitutes the basic unit of the overall multimodal data fusion algorithm. Each data unit contains an information kernel, a policy set, and a data pipeline set corresponding to the policy. The specific formula for the organizational structure of the data unit is expressed as follows: in, Represented as a data unit, Represented as information core, Represented as a set of strategies, Represented as a data pipeline set, information core It is a core component of the data unit, carrying key information and a set of strategies for achieving data intelligence. This includes input / output strategies, data standardization strategies, and mathematical logic operation strategies, as well as data pipelines. It serves as a channel for information transmission and communication, corresponding one-to-one with each strategy. New information kernels are generated through the association of data pipelines, realizing a hierarchical structure of data units. Then, initial data units are constructed and directly associated with the database, including the construction of initial information kernels, initial strategies, and initial data pipelines. The initial information kernel consists of field names and technical names extracted from the database, representing common attributes of the data. The initial strategy set includes basic and advanced strategies. Basic strategies involve data input / output, standardization, and basic mathematical logic operations. The initial data pipelines correspond one-to-one with each strategy, using strategy-defined data pipelines for data transmission and communication. The specific formula is as follows: in, Represented as the first One initial data unit, Represented as the first Each information core contains key information extracted from the database. Represented as based on information kernel The strategy function, Represented as a data pipeline function, it generates a data pipeline based on the information kernel and strategy. The index, representing a data unit, is defined by the following formula for the data standardization strategy: in, Represented as applied to the information core Data standardization strategy function, This is represented as a standardized operation. Represented as information core The mean, Represented as information core The standard deviation is then determined. Secondly, the data pipelines are randomly associated, and information kernels containing new information are continuously generated to construct higher-level data units. The initial data unit's data pipeline set is formed through all data pipelines of the initial data unit. Based on this, random association of data pipelines is performed to generate information kernels for binary combined data units. The specific formula is as follows: in, Represented as the set of random association results from the data pipeline. Represented as a data pipeline and The association results Represented as the first in the data pipeline One element, Represented as the first in the data pipeline One element, and Represented as the index of an element in the data pipeline. This is represented as a full-scale quantification operation. Then, by constructing a reward and punishment model and training the model by simulating human scoring of information cores, intelligent scoring and storage of the fused multimodal data are achieved. By automatically filtering out information cores with practical significance, high-quality data information is provided for the digital integrated application system based on intelligent highway management. The specific formula of the reward and punishment model is expressed as follows: in, This is represented as the loss function of the reward and punishment model. This represents the set of parameters for a reward and punishment model. Data samples represented as combined data units. It is represented as the set of training samples. Represented as a reward and punishment model based on the parameter set For data samples The prediction Represented as regularization parameters, Represented as data sample The actual label, Represented as the Sigmoid function, Represented as a logarithmic function, a task-driven intelligent information extraction method is finally constructed. Using actual task requests as information input, a topic model based on natural language understanding decomposes the task request into a set of topic titles for retrieving information kernels. Then, through random association in the data pipeline, a set of information kernels of combined data units related to the task request is generated. Through training the constructed reward and punishment model, intelligent ranking of the information kernels is achieved, storing the optimal matching information kernel for a given task request. The specific formula is expressed as follows: in, Represented as actual task requirements The loss function of the reward and punishment model, This is expressed as the actual task requirement. This is represented as an expectation operation, indicating that the expectation is calculated for all possible task requests and information verifications. Represented as the first Information core, Represented as the first Information core, and The index is represented as the information core. It is represented as a set of training samples, where each sample is a triple. The pairing of task requests and information cores and These are respectively represented as in a given task request Next, information core and The ranking in the sorted list is determined by minimizing the loss function. To optimize model parameters This enables the model to intelligently and autonomously extract information from the fused multimodal data.

6. The digital integrated application system based on intelligent highway maintenance according to claim 1, characterized in that, The data storage and management module ensures data security and availability by centrally storing and managing the collected multimodal data. It also supports the storage and retrieval of big data, provides query and analysis functions for historical data, and provides data support for the long-term management and planning of highways.

7. The digital integrated application system based on intelligent highway management and maintenance according to claim 1, characterized in that, The user monitoring and decision-making module provides an intuitive user interface and decision support functions, including data visualization tools and report generation functions, to help managers make correct decisions, while ensuring that highway managers can easily use the system to query data, view analysis results, and receive early warning information.

Images