Determination and evaluation method for optimal dosage of solid snow-melting agent

Abstract

The invention relates to the technical field of throwing of snow-melting agents, in particular to a method for measuring and evaluating the optimal dosage of a solid snow-melting agent, and the method comprises the steps: collecting environment data under different conditions through an environment sensor, recording the dosage of the snow-melting agent and an ice melting effect, and constructing a data set; training a snow-melting agent dosage prediction model based on a support vector machine (SVM) model, and estimating the required dosage according to environmental conditions; considering the snow removal effect and the environmental influence, and establishing a snow-melting agent dosage multi-objective optimization model; and the optimal snow-melting agent dosage is calculated by combining the prediction model and the optimization model, a dynamic optimization mechanism is established according to real-time feedback data, the model is continuously updated, and intelligent decision support is provided for snow removal of roads in winter.

Description

Technical Field

[0001] The present invention relates to the technical field of snow melting agent dispensing, and specifically to a method for determining and evaluating the optimal dosage of solid snow melting agent. Background Art

[0002] Solid snow melting agents play an important role in removing snow and ice on roads in winter. However, excessive use can have negative impacts on roads, the environment, and the ecosystem. With the improvement of environmental awareness, reducing the negative impact of snow melting agents on the environment has become an urgent need. Therefore, reasonably controlling the dosage of snow melting agents, which can not only achieve the effect of snow and ice melting but also reduce the negative impact on the environment, has become the focus of current research.

[0003] In the prior art, the use of solid snow melting agents is affected by conditions such as the temperature, humidity, and snow accumulation thickness of the actual road environment, which will cause great interference to the use of solid snow melting agents. Therefore, how to determine and evaluate the dosage of solid snow melting agents to determine the optimal dosage is the problem we need to solve.

[0004] For this reason, a method for determining and evaluating the optimal dosage of solid snow melting agent is proposed herein. Summary of the Invention

[0005] The present invention provides the following technical solution: A method for determining and evaluating the optimal dosage of solid snow melting agent, including:

[0006] Step 1: Obtain environmental data under different environmental conditions, record the dosage of snow melting agent and the ice melting effect, and construct a data set based on the dosage of snow melting agent, environmental data, and ice melting effect.

[0007] Step 2: Construct a snow melting agent dosage model based on the support vector machine (SVM) model, train the SVM model on the data set, and output the dosage of snow melting agent under given environmental data and snow accumulation amount through the snow melting agent dosage model.

[0008] Step 3: Based on the predicted result of the snow melting agent dosage output by the model, add constraint conditions of multi-objective optimization and environmental protection, construct a snow melting agent dosage optimization model, and define an optimization objective function.

[0009] Step 4: Based on the constructed snow melting agent dosage model and snow melting agent dosage optimization model, calculate the optimal dosage of snow melting agent under different conditions; establish a dynamic optimization mechanism for the dosage of snow melting agent, and update and optimize the snow melting agent dosage model according to the feedback data of the road snow removal effect.

[0010] Preferably, environmental data, including temperature, humidity, and snow accumulation thickness, are collected every [interval]; record the dosage of snow melting agent;

[0011] Evaluate the ice melting effect, define the evaluation index of the ice melting effect, consider the proportion of the ice melting area and the ice melting time, and the calculation formula is: where and are weight coefficients, satisfying;

[0012] Construct a data set, and combine the collected environmental data, the dosage of deicing agent and the evaluation index of the ice melting effect into a data set, where is the number of data samples.

[0013] Preferably, preprocess the data set, including removing outliers and data normalization; the removal of outliers includes: removing data with temperature, humidity and snow depth exceeding the measurement range;

[0014] The data normalization includes: performing min-max scaling on the temperature, humidity, snow depth and the dosage of deicing agent in the data set and mapping them into the interval [0, 1]. The formula is: where and are the minimum and maximum values of the feature respectively;

[0015] Use the Pearson correlation coefficient method to calculate the correlation coefficient between the features in the data set and the evaluation index of the ice melting effect, and select the features with the absolute value of the correlation coefficient greater than to obtain the final feature set, where.

[0016] Preferably, divide the data set into a training set and a test set according to the ratio of 7:3 to obtain a training data set, and use the training set to train the SVM model, where is the feature vector, is the number of data samples in the training data set, and is the corresponding dosage of deicing agent;

[0017] Use the SVM algorithm to train to obtain the SVM model, and the model parameters are selected through 5-fold cross-validation;

[0018] Calculate the dosage of deicing agent based on the SVM model. For the given environmental data, use the trained SVM model for prediction to obtain the dosage of deicing agent.

[0019] Preferably, evaluate the model performance on the test set, and calculate the root mean square error and the coefficient of determination;

[0020] Use the grid search method to optimize the penalty coefficient and the parameters of the RBF kernel function in the SVM model, select the parameter combination with the minimum value as the optimal parameter, and retrain the SVM model.

[0021] Preferably, define the objective function of the deicing agent dosage optimization model, including the objective of maximizing the ice melting effect and the objective of minimizing the environmental impact. The environmental impact index considers the residue concentration and duration. The calculation formula is: where and are weight coefficients, satisfying;

[0022] Set the constraint conditions of the deicing agent dosage optimization model, including the residue concentration limit, the upper limit of the deicing agent dosage and the upper limit of the ice melting time.

[0023] Preferably, the NSGA-II algorithm is used to transform the deicing agent dosage optimization model into a multi-objective optimization problem with two objectives: where represents the transformation of maximizing the deicing effect into a minimization problem;

[0024] The NSGA-II algorithm is used to solve the optimization problem, and combined with the deicing agent dosage predicted by the SVM model, the deicing agent dosage optimization problem is solved;

[0025] Set the algorithm parameters of the NSGA-II algorithm, including population size, crossover probability, mutation probability, and number of generations of evolution; obtain the optimal solution set, where is the number of optimal solutions.

[0026] Preferably, select the solution with the maximum deicing effect and the environmental impact not exceeding the threshold from the optimal solution set, which will be used as the optimal deicing agent dosage;

[0027] If there is no solution that meets the conditions, adjust the threshold until a solution that meets the conditions is found;

[0028] Collect dynamic feedback data. During the road snow removal process, collect feedback data on the deicing effect and environmental impact every time, including the proportion of ice melting area and residue concentration, which is recorded as the feedback data set, where is the number of feedback data samples.

[0029] Preferably, add the feedback data set to the original data set every time to obtain the updated data set; use to re-perform data preprocessing, feature engineering, and SVM model training to obtain the updated deicing agent dosage prediction model;

[0030] Establish a dynamic optimization mechanism for the deicing agent dosage. Every time, use the updated data set and the deicing agent dosage prediction model to reconstruct and solve the deicing agent dosage optimization model to obtain a new optimal solution set and the optimal deicing agent dosage;

[0031] According to the actual snow removal effect and environmental impact, dynamically adjust the weight coefficients,,, in the objective function, as well as the threshold in the constraint conditions to achieve the dynamic optimization of the deicing agent dosage.

[0032] A system for the determination and evaluation method of the optimal dosage of a solid deicing agent includes: a data acquisition module, a deicing agent dosage module, a multi-objective optimization module, and a dynamic optimization module;

[0033] The data acquisition module is used to obtain environmental data under different environmental conditions, record the deicing agent dosage and the ice melting effect, and construct the data into a data set according to the deicing agent dosage, environmental data, and ice melting effect;

[0034] The deicing agent dosage module constructs a deicing agent dosage model based on the support vector machine (SVM) model, performs SVM model training on the data set, and outputs the deicing agent dosage under the given environmental data and snow accumulation amount through the deicing agent dosage model;

[0035] The multi-objective optimization module constructs a deicing agent dosage optimization model and defines an optimization objective function by adding constraints of multi-objective optimization and environmental protection based on the predicted deicing agent dosage results output by the model.

[0036] The dynamic optimization module calculates the optimal deicing agent dosage under different conditions based on the constructed deicing agent dosage model and deicing agent dosage optimization model; establishes a dynamic optimization mechanism for deicing agent dosage, and updates and optimizes the deicing agent dosage model according to the feedback data on the road snow removal effect.

[0037] The beneficial effects of the present invention: By collecting environmental data and recording the deicing agent dosage and ice melting effect, the present invention constructs a high-quality data set, providing a reliable data basis for subsequent analysis.

[0038] The present invention trains a deicing agent dosage prediction model based on the SVM model, which can accurately estimate the required dosage according to environmental conditions and provide support for optimization decisions.

[0039] The present invention constructs a multi-objective optimization model for deicing agent dosage, taking into account the snow removal effect and environmental impact, achieving a balance in dosage decision-making and reducing the environmental burden.

[0040] The present invention realizes the dynamic optimization of deicing agent dosage, updates the model according to real-time feedback data, continuously provides effective snow removal decisions, and improves the snow removal efficiency.

[0041] The method of the present invention realizes the precise control and dynamic adjustment of deicing agent dosage through data-driven modeling and optimization, minimizes the environmental impact of deicing agent while ensuring road safety, and provides an intelligent, efficient and environmentally friendly winter road snow removal solution. Description of the Drawings

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them:

[0043] Figure 1 It is a flowchart of a method for determining and evaluating the optimal dosage of a solid deicing agent of the present invention;

[0044] Figure 2 It is a structural diagram of a system for determining and evaluating the optimal dosage of a solid deicing agent of the present invention. Detailed Embodiments

[0045] To make the above objects, features, and advantages of the present invention more understandable, the following provides a detailed description of the specific embodiments of the present invention in conjunction with the accompanying drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all 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 scope of protection of the present invention.

[0046] Embodiment 1

[0047] Referring to Figure 1 , the first embodiment of the present invention provides a method for determining and evaluating the optimal dosage of a solid snow melting agent.

[0048] Step 1: Obtain environmental data under different environmental conditions, record the dosage of the snow melting agent and the ice melting effect, and construct a data set based on the dosage of the snow melting agent, environmental data, and ice melting effect.

[0049] Using a temperature and humidity sensor and a snow depth sensor device, collect environmental data every Δt, including temperature T (-50°C ≤ T ≤ 50°C, accuracy ±0.5°C), humidity H (0% ≤ H ≤ 100, accuracy ±2), and snow accumulation thickness S (0 cm ≤ S ≤ 100 cm, accuracy ±1 cm); record the dosage of the snow melting agent M (0 g / m 2 ≤ M ≤ 1000 g / m 2 , accuracy ±1 g / m 2 ).

[0050] Evaluate the ice melting effect, define the ice melting effect evaluation index E, considering the proportion of the ice melting area R a (0% ≤ R a ≤ 100, accuracy ±1) and the ice melting time t m (0 min ≤ t m ≤ 1440 min, accuracy ±1 min), and the calculation formula is: Where α and β are weight coefficients, satisfying α + β = 1.

[0051] Construct a data set, and combine the collected environmental data, the dosage of the snow melting agent, and the ice melting effect evaluation index into a data set Where N is the number of data samples.

[0052] Preprocess the data set D, including removing outliers and data normalization; the removal of outliers includes: removing data where the temperature, humidity, and snow accumulation thickness exceed the measurement range.

[0053] The data normalization includes: performing min-max scaling on the temperature, humidity, snow accumulation thickness, and dosage of the snow melting agent in the data set D, and mapping them into the interval [0, 1], and the formula is: Where xmin and x max are the minimum and maximum values of the feature, respectively.

[0054] Use the Pearson correlation coefficient method to calculate the correlation coefficient between the features in the dataset and the ice melting effect evaluation index E, and select the features with the absolute value of the correlation coefficient greater than θ to obtain the final feature set F = f 1 , f 2 ,..., f k , where k ≤ 4.

[0055] Step 1 collects data through environmental sensors, performs preprocessing and feature selection, and constructs a high-quality dataset containing temperature, humidity, snow depth, deicer dosage, and ice melting effect, providing a reliable data basis for subsequent model training.

[0056] Step 2: Build a deicer dosage model based on the support vector machine (SVM) model, train the SVM model on the dataset, and output the deicer dosage under the given environmental data and snow accumulation.

[0057] Divide the dataset D into a training set D train and a test set D test in a ratio of 7:3 to obtain the training dataset Use the training set to train the SVM model, where x i is the feature vector, N train is the number of data samples in the training dataset, and y i is the corresponding deicer dosage M i .

[0058] Use the SVM algorithm to train D train to obtain the SVM model f SVM , and the model parameters are selected through 5-fold cross-validation.

[0059] Calculate the deicer dosage based on the SVM model. For the given environmental data x new , use the trained SVM model f SVM to make a prediction and obtain the deicer dosage

[0060] Evaluate the model performance on the test set and calculate the root mean square error and the coefficient of determination where is the average value of y i , and N test is the number of data samples in the test set.

[0061] Using the grid search method, optimize the penalty coefficient κ and the RBF kernel function parameter γ in the SVM model, taking values at intervals of powers of 10 within the search range, and select the parameter combination with the minimum RMSE as the optimal parameters to retrain the SVM model.

[0062] Step 2: Use the SVM algorithm to train the deicing agent dosage prediction model, optimize the model parameters through cross-validation and grid search, and improve the prediction accuracy of the model. This model can accurately predict the required deicing agent dosage based on environmental data and provide support for optimization decisions.

[0063] Step 3: Based on the deicing agent dosage prediction results output by the model, add constraints on multi-objective optimization and environmental protection to construct a deicing agent dosage optimization model and define the optimization objective function.

[0064] Define the objective function of the deicing agent dosage optimization model, including the objective of maximizing the deicing effect maxE and the objective of minimizing the environmental impact minI, where the environmental impact index I considers the residue concentration C (0mg / L ≤ C ≤ 100mg / L, accuracy ±0.1mg / L) and the duration t r (0h ≤ t r ≤ 720h, accuracy ±1h), and the calculation formula is: where γ and δ are weight coefficients, satisfying γ + δ = 1.

[0065] Set the constraint conditions of the deicing agent dosage optimization model, including the residue concentration limit maxC, the upper limit of the deicing agent dosage maxM, and the upper limit of the ice melting time maxt m .

[0066] Adopt the NSGA-II algorithm to transform the deicing agent dosage optimization model into a multi-objective optimization problem with two objectives: min(-E, I), where -E represents transforming the maximization of the deicing effect into a minimization problem.

[0067] Adopt the NSGA-II algorithm to solve the optimization problem, and combine the deicing agent dosage predicted by the SVM model to solve the deicing agent dosage optimization problem.

[0068] Set the algorithm parameters of the NSGA-II algorithm, including the population size N pop , the crossover probability p c , the mutation probability p m , the number of generations of evolution N gen ; obtain the optimal solution set where N p is the number of optimal solutions.

[0069] Step 3 constructs a multi-objective optimization model for the amount of snow melting agent, considering the snow melting effect and environmental impact, and sets reasonable constraints. The NSGA-II algorithm is used to solve the model, obtaining the Pareto optimal solution set for the amount of snow melting agent and achieving the balance between the snow melting effect and environmental impact.

[0070] Step 4: Based on the constructed model for the amount of snow melting agent and the optimization model for the amount of snow melting agent, calculate the optimal amount of snow melting agent under different conditions; establish a dynamic optimization mechanism for the amount of snow melting agent, and update and optimize the model for the amount of snow melting agent according to the feedback data on the road snow removal effect.

[0071] Select the solution (M, E, I) from the optimal solution set P with the maximum snow melting effect E and the environmental impact I not exceeding the threshold I th and take (M, E, I) * as the optimal amount of snow melting agent. * If there is no solution that meets the conditions, adjust the threshold I

[0072] until a solution that meets the conditions is found; th Collect dynamic feedback data. During the road snow removal process, collect the feedback data on the snow melting effect and environmental impact every Δt

[0073] , including the proportion of ice melting area R f and the residue concentration C, denoted as the feedback data set a where N is the number of feedback data samples. f Every Δt

[0074] , add the feedback data set D up to the original data set D to obtain the updated data set D f = D ∪ D up ; use D f to re-perform data preprocessing, feature engineering, and SVM model training to obtain the updated prediction model f up for the amount of snow melting agent; SVM,up ;

[0075] Establish a dynamic optimization mechanism for the amount of snow melting agent. Every Δt op , use the updated data set D up and the prediction model f SVM,up for the amount of snow melting agent to re-construct and solve the optimization model for the amount of snow melting agent, obtaining the new optimal solution set P up and the optimal amount of snow melting agent

[0076] According to the actual snow removal effect and environmental impact, dynamically adjust the weight coefficients α, β, γ, δ in the objective function and the threshold I in the constraints th to obtain the dynamic optimization result.

[0077] In step 4, the optimal snow melting agent dosage is selected according to the optimal solution set, and a dynamic optimization mechanism is established. By continuously collecting feedback data, regularly updating the data set and the model, and dynamically adjusting the weights of the objective function and the constraint thresholds, the optimization results can adapt to environmental changes and provide continuous and effective snow removal decision support.

[0078] The technical solution of this embodiment constructs an efficient, accurate and environmentally friendly dynamic optimization method for the dosage of snow melting agent. This method collects data in real time through environmental sensors, uses the SVM model to predict the dosage of snow melting agent, and combines the NSGA-II algorithm for multi-objective optimization to minimize the environmental impact while ensuring the snow removal effect. The model is dynamically updated and optimized according to the actual feedback data to enable it to adapt to changes in environmental conditions and provide intelligent decision support for winter road snow removal.

[0079] Embodiment 2

[0080] Refer to Figure 2 , the second embodiment of the present invention provides a measurement and evaluation system for the optimal dosage of solid snow melting agent.

[0081] The system includes: a data acquisition module, a snow melting agent dosage module, a multi-objective optimization module, and a dynamic optimization module.

[0082] The data acquisition module is used to obtain environmental data under different environmental conditions, record the dosage of snow melting agent and the ice melting effect, and construct a data set based on the dosage of snow melting agent, environmental data, and ice melting effect.

[0083] The snow melting agent dosage module constructs a snow melting agent dosage model based on the support vector machine (SVM) model, trains the SVM model on the data set, and outputs the dosage of snow melting agent under given environmental data and snow accumulation.

[0084] The multi-objective optimization module, based on the predicted results of the snow melting agent dosage output by the model, adds constraint conditions for multi-objective optimization and environmental protection, constructs a snow melting agent dosage optimization model, and defines the optimization objective function.

[0085] The dynamic optimization module, based on the constructed snow melting agent dosage model and snow melting agent dosage optimization model, calculates the optimal snow melting agent dosage under different conditions; establishes a dynamic optimization mechanism for the snow melting agent dosage, and updates and optimizes the snow melting agent dosage model according to the feedback data of the road snow removal effect.

[0086] Embodiment 3

[0087] The third embodiment of the present invention is different from the previous embodiment in that:

[0088] If the above-mentioned functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs.

[0089] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a predefined sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0090] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection part with one or more wirings (electronic device), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, then editing, interpreting, or otherwise processing it as appropriate, and then storing it in a computer memory.

[0091] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0092] In addition, in order to provide a concise description of the exemplary embodiments, all features of the actual embodiments may not be described (i.e., those features that are not relevant to the currently contemplated best mode of carrying out the present invention, or those features that are not relevant to implementing the present invention).

[0093] It should be understood that in the development of any actual implementation, in any engineering or design project, a large number of specific implementation decisions may be made. Such development efforts may be complex and time-consuming, but for those of ordinary skill in the art who benefit from this disclosure, without undue experimentation, such development efforts will be a routine task of design, manufacturing, and production.

[0094] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A method for determining and evaluating the optimal dosage of a solid deicing agent, characterized in that: include: Step 1: Obtain environmental data under different environmental conditions, record the amount of deicing agent used and the ice melting effect, and construct the data into a data set based on the amount of deicing agent used, environmental data, and ice melting effect; Step 2: Construct a deicing agent dosage model based on the support vector machine (SVM) model, train the SVM model on the data set, and output the deicing agent dosage under given environmental data and snow accumulation through the deicing agent dosage model; Step 3: Based on the deicing agent dosage prediction results output by the model, add multi-objective optimization and environmental protection constraints, build a deicing agent dosage optimization model, and define the optimization objective function; Step 4: Based on the constructed deicing agent dosage model and deicing agent dosage optimization model, calculate the optimal deicing agent dosage under different conditions; establish a dynamic optimization mechanism for deicing agent dosage, and update and optimize the deicing agent dosage model based on feedback data on road snow removal effects.

2. The method for determining and evaluating the optimal dosage of a solid deicing agent according to claim 1, characterized in that: Collect environmental data every Δt, including temperature T, humidity H and snow thickness S; record the amount of snow melting agent M; Evaluate the ice melting effect, define the ice melting effect evaluation index E, and consider the ice melting area ratio R a and melting time t m , the calculation formula is: Where α and β are weight coefficients, satisfying α+β=1; Construct a data set by combining the collected environmental data, deicing agent dosage, and deicing effect evaluation indicators into a data set Where N is the number of data samples.

3. The method for determining and evaluating the optimal dosage of a solid deicing agent according to claim 2, characterized in that: Preprocess the data set D, including removing outliers and normalizing the data; The removal of abnormal values ​​includes: removing data of temperature, humidity and snow thickness that exceed the measurement range; The data normalization includes: performing minimum-maximum scaling on the temperature, humidity, snow thickness and deicing agent dosage in the data set D and mapping them to the interval [0, 1], and the formula is: where x min and x max are the minimum and maximum values ​​of the feature respectively; The Pearson correlation coefficient method is used to calculate the correlation coefficient between the features in the data set and the ice melting effect evaluation index E. The features with an absolute value of the correlation coefficient greater than θ are selected to obtain the final feature set F = f1, f2, ..., f k , where k≤4.

4. The method for determining and evaluating the optimal dosage of a solid deicing agent according to claim 3, characterized in that: The dataset D is divided into training set D according to the ratio of 7:

3. train and the test set D test , get the training data set Use the training set to train the SVM model, where x i is the feature vector, N train is the number of data samples in the training data set, y i is the corresponding amount of snow melting agent M i ; Using SVM algorithm to train Train and get the SVM model f SVM , model parameters were selected by 5-fold cross validation; Based on the SVM model, the amount of deicing agent is calculated for the given environmental data x new , using the trained SVM model f SVM Make a prediction to get the amount of deicing agent used 5. The method for determining and evaluating the optimal dosage of a solid deicing agent according to claim 4, characterized in that: Evaluate model performance on the test set and calculate the root mean square error and coefficient of determination; The grid search method is used to optimize the penalty coefficient κ and RBF kernel function parameter γ in the SVM model, and the parameter combination with the smallest RMSE is selected as the optimal parameter to retrain the SVM model.

6. The method for determining and evaluating the optimal amount of a solid deicing agent according to claim 5, characterized in that: Define the objective function of the deicing agent dosage optimization model, including the snow melting effect maximization target maxE and the environmental impact minimization target minI, where the environmental impact index I takes into account the residue concentration C and duration t r , the calculation formula is: Where γ and δ are weight coefficients, satisfying γ+δ=1; Set constraints for the deicing agent dosage optimization model, including the residue concentration limit maxC, the deicing agent dosage upper limit maxM, and the ice melting time upper limit maxt m .

7. The method for determining and evaluating the optimal dosage of a solid deicing agent according to claim 6, characterized in that: The NSGA-II algorithm is used to transform the deicing agent dosage optimization model into a multi-objective optimization problem with two objectives: min(-E,I), where -E represents the transformation of the deicing effect maximization into a minimization problem; The NSGA-II algorithm is used to solve the optimization problem, and the deicing agent dosage prediction by the SVM model is combined to solve the deicing agent dosage optimization problem; Set the NSGA-II algorithm parameters, including the population size N pop , crossover probability p c , mutation probability p m , evolutionary algebra N gen ; Get the optimal solution set Where N p is the number of optimal solutions.

8. The method for determining and evaluating the optimal amount of solid deicing agent according to claim 7, characterized in that: Select the optimal solution set P with the maximum snowmelt effect E and environmental impact I not exceeding the threshold I th The solution (M, E, I * ), convert (M,E,I * ) as the optimal amount of snow melting agent; If there is no solution that satisfies the conditions, adjust the threshold I th , until a solution that satisfies the conditions is found; Collect dynamic feedback data, during the road snow removal process, every Δt f Collect feedback data on snow melting effect and environmental impact, including the proportion of ice melting area R a and residue concentration C, denoted as feedback data set Where N f is the number of feedback data samples.

9. The method for determining and evaluating the optimal amount of solid deicing agent according to claim 8, characterized in that: Every Δt up The feedback dataset D f Add to the original dataset D to get the updated dataset D up =D∪D f ; Use D up Re-perform data preprocessing, feature engineering and SVM model training to obtain the updated deicing agent dosage prediction model f SVM,up ; Establish a dynamic optimization mechanism for deicing agent dosage, and op Using the updated dataset D up and deicing agent dosage prediction model SVM,up , reconstruct and solve the deicing agent dosage optimization model, and obtain a new optimal solution set P up and optimal de-icing agent dosage According to the actual snow removal effect and environmental impact, the weight coefficients α, β, γ, δ in the objective function and the threshold I in the constraint condition are dynamically adjusted. th , to achieve dynamic optimization of snow-melting agent dosage.

10. A system for determining and evaluating the optimal amount of solid deicing agent according to any one of claims 1 to 9, characterized in that: include: Data acquisition module, deicing agent dosage module, multi-objective optimization module and dynamic optimization module; The data acquisition module is used to acquire environmental data under different environmental conditions, record the amount of deicing agent used and the ice-melting effect, and construct the data into a data set according to the amount of deicing agent used, the environmental data and the ice-melting effect; The deicing agent dosage module constructs a deicing agent dosage model based on a support vector machine (SVM) model, performs SVM model training on a data set, and outputs deicing agent dosage under given environmental data and snow accumulation through the deicing agent dosage model; The multi-objective optimization module, based on the deicing agent dosage prediction result output by the model, adds multi-objective optimization and environmental protection constraints, builds a deicing agent dosage optimization model, and defines the optimization objective function; The dynamic optimization module calculates the optimal de-icing agent dosage under different conditions based on the constructed de-icing agent dosage model and de-icing agent dosage optimization model; establishes a dynamic optimization mechanism for de-icing agent dosage, and updates and optimizes the de-icing agent dosage model based on feedback data on road snow removal effects.

Images