Prediction optimization and method for rural household garbage clearance opportunity

Abstract

The invention provides a rural household garbage clearance opportunity prediction and optimization method, and the method comprises the specific steps: carrying out the clearance of rural household garbage through a garbage clearance vehicle, and obtaining historical clearance data; preprocessing the historical clearance data; building a garbage clearance prediction model by adopting a support vector machine, and obtaining optimal parameters of the garbage clearance prediction model by adopting a sparrow search algorithm; based on the garbage clearance prediction model and the historical clearance data, the garbage generation amount in the future time period is predicted; and determining a garbage clearance scheme based on the garbage generation amount in the future time period. According to the method, historical clearance data is collected, machine learning and data analysis means are utilized, a garbage collection site is used as a unit, a prediction model is established by adopting a support vector machine, and parameter optimization is performed on the support vector machine by adopting a sparrow search algorithm, so that accurate prediction of the amount of garbage generated at the collection site in the future is realized; and the opportunity suitable for garbage collection is further deduced.

Description

Technical Field

[0001] The present invention relates to the technical field of garbage disposal, and in particular to a method for predicting and optimizing the timing of clearing and transporting rural domestic garbage. Background Art

[0002] In order to effectively improve the living environment in rural areas, many villages have set up garbage collection points. The garbage collection at these collection points is usually carried out according to a certain cycle.

[0003] Due to the different number of households in rural areas, the garbage generation situation in rural areas is more complicated than that in cities. Some rural areas have more households, so the speed of garbage generation is generally faster, and the amount of garbage generated is also correspondingly larger. Some rural areas have fewer households, and not much garbage is generated for a long time. If garbage is removed and transported according to a fixed cycle, some collection points will have overflowing garbage but not removed in time, and some collection points will have no garbage but have been removed. This will result in untimely collection and transportation or waste of collection and transportation costs. Therefore, a reasonable time for removal is particularly important.

[0004] The advantage of garbage collection according to a fixed cycle is that it is simple and easy to manage, and does not require complex real-time monitoring and prediction. However, the disadvantages are also obvious. First, fixed-cycle collection may lead to waste of resources and increase collection costs. If less garbage is generated in certain cycles, the collection personnel and vehicles may waste time and resources. Secondly, it is easy to cause garbage accumulation. Some collection points generate garbage at a faster rate, and garbage is easy to accumulate in the garbage can or even overflow the garbage can, which may cause environmental and sanitation problems. Thirdly, fixed-cycle collection is not suitable for fluctuations in demand. The amount of garbage generated may fluctuate due to seasons, special events, climate and other reasons. Fixed-cycle collection cannot adapt well to these fluctuations, which may lead to insufficient or excessive collection. Summary of the invention

[0005] In view of this, the present invention proposes a method for predicting and optimizing the timing of rural domestic waste removal to solve the problems existing in the above-mentioned prior art.

[0006] To achieve the above object, the present invention proposes a method for predicting and optimizing the timing of rural domestic garbage removal, comprising:

[0007] Use garbage trucks to remove rural domestic garbage and obtain historical removal data;

[0008] Preprocessing the historical removal data;

[0009] A support vector machine is used to construct a garbage removal prediction model, and a sparrow search algorithm is used to obtain the optimal parameters of the garbage removal prediction model;

[0010] Predicting the amount of garbage generated in a future period based on the garbage removal prediction model and the historical removal data;

[0011] A garbage removal plan is determined based on the amount of garbage generated in the future period.

[0012] Optionally, the historical removal data includes: garbage collection point information, removal time, removal vehicle information, garbage weight, and weather information; and the process of preprocessing the historical removal data includes:

[0013] Using principal component analysis method to perform dimension reduction and filtering on the historical transportation data;

[0014] A consistent approach is used to quality control, screen, fill in gaps and split the historical removal data.

[0015] Optionally, the process of constructing a garbage removal prediction model using a support vector machine includes:

[0016] Based on a support vector machine, an input sample in a low-dimensional space is mapped to a high-dimensional vector space through a nonlinear mapping function, and a linear regression function is constructed in the high-dimensional vector space;

[0017] Based on the principle of structural risk minimization, the regression problem of the support vector machine is expressed as a mathematical optimization problem;

[0018] The Lagrange multiplier is introduced to transform the solution of the mathematical optimization problem into a saddle point for solving the Lagrange equation. After calculating the partial derivatives of the equation and applying the duality theorem, the model optimization problem is obtained.

[0019] Optionally, the mathematical optimization problem is as follows:

[0020]

[0021]

[0022] In the formula, i = 1, 2, ..., n, n is the number of samples, C is the penalty factor, ε is the loss function, and y i =f(x i ),ξ i and i * is the slack variable.

[0023] Optionally, the model optimization problem is as follows:

[0024]

[0025] In the formula, α i ,α i *is a Lagrange multiplier, and α i >0,α i * >0, represents the kernel function, i,j=1,2,…,n.

[0026] Optionally, the process of using the sparrow search algorithm to obtain the optimal parameters of the garbage removal prediction model includes:

[0027] The sparrow search algorithm seeks the optimal solution of the model parameters by simulating the process of sparrows searching for food, including an initial stage, an update stage and a determination stage;

[0028] In the initial stage, the initial position of the sparrow population is randomly generated, and the fitness of several individuals in the sparrow population is determined. The sparrow population includes discoverers, joiners and early warnings; the discoverer is an individual with high fitness; the discoverer updates its position according to formula 1

[0029] In the update phase, the optimal position is obtained by the discoverer, and the joiner competes with the discoverer for the optimal position. If the competition is successful, the joiner moves to the optimal position; if the competition fails, the joiner updates its position according to Formula 2;

[0030] The threat is discovered by the early warning device. When there is a threat, all sparrows update their positions based on Formula 3. When there is no threat, the discoverer enters a wide search mode.

[0031] After all sparrows have updated their positions, the determination phase is entered, and the current position of each individual sparrow is compared with the position of the last repetition. If the new position is better than the previous position and the best position is saved, the update is performed;

[0032] Repeat the iterative update of the position, and when the number of repetitions reaches the maximum number, output the optimal parameters.

[0033] Optionally, the formula 1 is as follows:

[0034]

[0035] Where: iter max is the constant with the highest iteration number, t is the current iteration, j = 1, 2, ..., d, Denote the next value of the j-th sparrow in the t-th iteration; α is a random number between 0 and 1; R2 is a number in the range of 0 to 1, and ST is a value from 0.5 to 1.0; Q is a random number based on the normal distribution; L represents a 1×d matrix where each element is 1; if R2≥ST, it means that some sparrows have detected a predator and all sparrows must quickly fly to other safe areas; when R2<ST, it means there is no predator around and the discoverers enter the extensive search mode.

[0036] Optionally, Formula Two is as follows:

[0037]

[0038] where i, j = 1, 2, …, n; t represents the number of iterative updates, denote the latest position of the joiner, denote the worst position.

[0039] Optionally, Formula Three is as follows:

[0040]

[0041] In the formula: X best is the current global optimal position; ε is a small constant to avoid zero division error; β is the control parameter of the step size and the normal distribution of random numbers with a mean of 0 and a variance of 1; f g and f w are the current best and worst overall suitability values respectively; K is a random number between 1 and -1; f i is the current value of the sparrow; f i = f g indicates that the sparrows in the middle of the population are aware of the danger; if f i > f g , then this individual is at the edge of the population.

[0042] Optionally, the process of determining the garbage removal plan based on the garbage generation amount in the future period includes:

[0043] Construct the maximum garbage accumulation amount. When the sum of the garbage generation amounts in the future period output by the model does not reach the maximum garbage accumulation amount, no garbage removal is carried out; when the sum of the garbage generation amounts in the future period output by the model reaches the maximum garbage accumulation amount, garbage removal is carried out.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0045] The present invention collects historical collection data, utilizes machine learning and data analysis methods, takes garbage collection sites as units, uses support vector machines to build a prediction model, and uses the sparrow search algorithm to optimize the parameters of the support vector machine, so as to achieve accurate prediction of the future amount of garbage generated at the collection point, and then infer the time suitable for garbage collection, thereby providing a reference basis for more accurate and effective rural garbage collection. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0047] Figure 1 The figure is a flow chart of a method for predicting and optimizing the timing of rural domestic waste removal in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to be able to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0049] This embodiment proposes a method for predicting and optimizing the timing of rural domestic garbage removal. Figure 1 As shown, the following steps are included:

[0050] Use garbage trucks to remove rural domestic garbage and obtain historical removal data;

[0051] Preprocessing the historical removal data;

[0052] A support vector machine is used to construct a garbage removal prediction model, and a sparrow search algorithm is used to obtain the optimal parameters of the garbage removal prediction model;

[0053] Predicting the amount of garbage generated in a future period based on the garbage removal prediction model and the historical removal data;

[0054] A garbage removal plan is determined based on the amount of garbage generated in the future period.

[0055] This embodiment obtains the historical data of garbage collection trucks collecting rural biochemical garbage, and after collecting statistics on the collection point information, collection time, collection truck information, garbage weight, weather information, etc., the principal component analysis method is used to reduce the dimension and filter the statistical data. At the same time, the consistency method is used to control the quality, screen, fill in the gaps and split the statistical data, and finally obtain the historical garbage collection data for model input.

[0056] The present embodiment adopts support vector machine SVM as the basic network structure of the model. SVM is a supervised machine learning classifier model based on hyperplane data separation. The concept of SVM was originally used to solve classification problems, but was later widely used in regression problems, such as function approximation and prediction. It constructs linear and nonlinear decision boundaries (hyperplanes) by finding support vectors and optimizing distance formulas. For nonlinearly separable data, SVM introduces the concept of kernel functions (polynomials and radial basis functions) to convert data into another form that can be separated by hyperplanes. The main advantage of SVM lies in its ability to handle high-dimensional data and nonlinear relationships. It can find the optimal hyperplane in high-dimensional space, thereby realizing classification or regression tasks. In addition, SVM has strong generalization ability and robustness to outliers.

[0057] For a given data set, SVM transforms the i-th input sample x in the low-dimensional space i Through nonlinear mapping function Mapped to a high-dimensional vector space. The linear regression function established in the high-dimensional space is simplified to:

[0058] f(x)=w·φ(x)+b,(x∈R d ,b∈R)

[0059] Where w represents the weight, b is the parameter of the bias, f(x) represents the predicted value, x is the input vector in the sample space, R represents the real number domain, and d represents the dimension of the sample space.

[0060] Applying the principle of structural risk minimization, the support vector regression problem is replaced by a constrained mathematical optimization problem. The fitting error is taken into account and two slack variables ξ are added. i and i * The optimization function and constraints are as follows:

[0061]

[0062] In the formula, i = 1, 2, ..., n, n is the number of samples, C is the penalty factor, ε is the loss function, and y i =f(x i ).

[0063] By introducing Lagrange multipliers, the solution of the equation can be transformed into solving the saddle point of the Lagrange equation. After calculating the partial derivatives of the equation and applying the duality theorem, the final optimization problem is as follows:

[0064]

[0065] In the formula, α i ,α i * is a Lagrange multiplier, and α i >0,α i * >0, represents the kernel function, i,j=1,2,…,n.

[0066] Furthermore, the nonlinear function is expressed as:

[0067]

[0068] The radial basis function is considered an ideal classification function due to its wide convergence region and high resolution. In this study, the radial basis function is selected as the kernel function, which is defined as:

[0069]

[0070] In the formula, d is the width of the radial basis kernel function.

[0071] In addition, in order to avoid the slow convergence speed and low simulation accuracy caused by the differences in the units, dimensions, and positive and negative properties of the input variables, each variable needs to be normalized and preprocessed so that its numerical value is linearly mapped within the range of 0 to 1. The calculation method is shown in the following formula:

[0072]

[0073] Where: X * is the value after standardized calculation; X is the initial result of the variable; X max , X min are the maximum and minimum values ​​of the original data of the variable respectively.

[0074] The application of radial basis functions helps to show the relationship between the original input and output space. However, in traditional SVM, the penalty factor C and parameter d are selected empirically. In cases where the input is diverse and complex, this approach increases the randomness of the prediction and leads to unreliable results. To address this problem, the sparrow search algorithm is used to select the optimal parameters of the support vector machine.

[0075] The Sparrow Search Algorithm (SSA) is a swarm intelligence-based optimization algorithm used for solving complex optimization problems. It draws on the foraging lifestyle and anti-predation behavior of sparrows and searches for the optimal solution by simulating the process of sparrows searching for food. This algorithm has fewer parameters, stronger search capabilities, and faster performance. The main phases of SSA can be explained as follows:

[0076] i) Create and initialize the solution, randomly generate the initial positions of the sparrow population:

[0077]

[0078] where: n is the number of sparrows, and d is the dimension of the selected variables.

[0079] Use formula (2) to judge the fitness of individuals for the following procedures:

[0080]

[0081] ⅱ) In the Sparrow Search Algorithm, discoverers with higher fitness values are given priority. Since discoverers are responsible for finding food and guiding the movement of the entire group, they can search for food over a wider range compared to the positions of joiners. In each iteration, discoverers update their states using formula (3).

[0082]

[0083] where: iter max is a constant with the highest number of iterations, t is the current iteration, j = 1, 2, …, d, represents the next value of the j-th sparrow in the t-th iteration. α is a random number between 0 and 1. R2 (alarm value) is a number in the range of 0 to 1, and ST (safety threshold) is a value between 0.5 and 1.0. Q is a random number based on the normal distribution. L represents a 1×d matrix where each element is 1. If R2 ≥ ST, it means that some sparrows have detected a predator, and all sparrows must quickly fly to other safe areas. When R2 < ST, it means there is no predator around, and the discoverers enter the extensive search mode.

[0084] For joiners, they must follow formula (3) and formula (4). As mentioned before, some joiners closely monitor the discoverers. When they learn that the discoverers have found food, they will leave their current locations to compete for the food. If they win, they can eat immediately, and at this time they follow formula 3 for position update; otherwise, they will adapt to formula (4):

[0085]

[0086] ⅲ) After updating the position of the entire population, select a few sparrows as scouts (explorers) to be responsible for identification and early warning. They usually account for 10% to 20% of the population. Update their positions according to formula (5):

[0087]

[0088] Where: X best is the current global optimal position. ε is a small constant to avoid zero division errors. β is the control parameter for the step size and the random number normal distribution with mean 0 and variance 1. g and f w are the current best and worst overall suitability values, respectively. K is a random number between 1 and -1. i is the current value of sparrow. i =f g It means that the sparrows in the middle of the population are aware of the danger. i >f g , then the individual is at the edge of the population.

[0089] iv) Compare the current position of each individual with the position of the last repetition. If the new position is better than the previous one and the best position is saved, it will be updated.

[0090] ⅴ) If the number of repetitions is less than the maximum number, go to step 2. Otherwise, the algorithm stops and the optimal solution is obtained.

[0091] The support vector machine SVM model with optimal parameters is obtained through the above iterative optimization process. The normalized statistical data is input into the model to predict the amount of garbage generated in future time periods. At this time, the garbage generation output by the model is counted, and the total amount of garbage that the site can accommodate is set for each garbage collection point. By adding up the garbage generation and comparing it with the total amount of garbage, when the garbage generation of a certain site reaches the total amount of garbage, garbage removal measures can be taken for the site, and the garbage removal times of different sites are counted to generate a garbage removal plan.

[0092] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can still modify the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or perform equivalent replacements on some of the technical features thereof; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. They should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for predicting and optimizing the timing of rural domestic waste removal, characterized in that: The following steps are involved: Use garbage trucks to remove rural domestic garbage and obtain historical removal data; Preprocessing the historical removal data; A support vector machine is used to construct a garbage removal prediction model, and a sparrow search algorithm is used to obtain the optimal parameters of the garbage removal prediction model; Predicting the amount of garbage generated in a future period based on the garbage removal prediction model and the historical removal data; A garbage removal plan is determined based on the amount of garbage generated in the future period.

2. The method for predicting and optimizing the timing of rural domestic waste removal according to claim 1, characterized in that: The historical removal data includes: garbage collection point information, removal time, removal vehicle information, garbage weight, and weather information; the process of preprocessing the historical removal data includes: Using principal component analysis method to perform dimension reduction and filtering on the historical transportation data; A consistent approach is used to quality control, screen, fill in gaps and split the historical removal data.

3. The method for predicting and optimizing the timing of rural domestic waste removal according to claim 1, characterized in that: The process of building a garbage removal prediction model using support vector machines includes: Based on a support vector machine, an input sample in a low-dimensional space is mapped to a high-dimensional vector space through a nonlinear mapping function, and a linear regression function is constructed in the high-dimensional vector space; Based on the principle of structural risk minimization, the regression problem of the support vector machine is expressed as a mathematical optimization problem; The Lagrange multiplier is introduced to transform the solution of the mathematical optimization problem into a saddle point for solving the Lagrange equation. After calculating the partial derivatives of the equation and applying the duality theorem, the model optimization problem is obtained.

4. The method for predicting and optimizing the timing of rural domestic waste removal according to claim 3 is characterized in that: The mathematical optimization problem is as follows: In the formula, i = 1, 2, ..., n, n is the number of samples, C is the penalty factor, ε is the loss function, and y i =f(x i ),ξ i and i * is the slack variable.

5. The method for predicting and optimizing the timing of rural domestic waste removal according to claim 3, characterized in that: The model optimization problem is as follows: In the formula, α i ,α i * is a Lagrange multiplier, and α i >0,α i * >0, represents the kernel function, i,j=1,2,…,n.

6. The method for predicting and optimizing the timing of rural domestic waste removal according to claim 1, characterized in that: The process of using the sparrow search algorithm to obtain the optimal parameters of the garbage removal prediction model includes: The sparrow search algorithm seeks the optimal solution of the model parameters by simulating the process of sparrows searching for food, including an initial stage, an update stage and a determination stage; In the initial stage, the initial position of the sparrow population is randomly generated, and the fitness of several individuals in the sparrow population is determined. The sparrow population includes discoverers, joiners and early warnings; the discoverer is an individual with high fitness; the discoverer updates its position according to formula 1 In the update phase, the optimal position is obtained by the discoverer, and the joiner competes with the discoverer for the optimal position. If the competition is successful, the joiner moves to the optimal position; if the competition fails, the joiner updates its position according to Formula 2; The threat is discovered by the early warning device. When there is a threat, all sparrows update their positions based on Formula 3. When there is no threat, the discoverer enters a wide search mode. After all sparrows have updated their positions, the determination phase is entered, and the current position of each individual sparrow is compared with the position of the last repetition. If the new position is better than the previous position and the best position is saved, the update is performed; Repeat the iterative update of the position, and when the number of repetitions reaches the maximum number, output the optimal parameters.

7. The method for predicting and optimizing the timing of rural domestic waste removal according to claim 6, characterized in that: The formula 1 is as follows: where: iter max is a constant with the highest number of iterations, t is the current iteration, j = 1, 2, …, d, represents the next value of the j-th sparrow in the t-th iteration; α is a random number between 0 and 1; R2 is a number in the range of 0 to 1, ST is a value from 0.5 to 1.0; Q is a random number based on the normal distribution; L represents a 1×d matrix where each element is 1; if R2≥ST, it means that some sparrows have detected a predator and all sparrows must quickly fly to other safe areas; when R2<ST, it means that there is no predator around and the discoverers enter the extensive search mode.

8. The method for predicting and optimizing the timing of rural domestic waste removal according to claim 6, characterized in that: The formula 2 is as follows: Where i,j=1,2,…,n; t represents the number of iterative updates. Indicates the latest location of the joiner. Indicates the worst position.

9. The method for predicting and optimizing the timing of rural domestic waste removal according to claim 6, characterized in that: The formula three is as follows: Where: X best is the current global optimal position; ε is a small constant to avoid zero division errors; β is the control parameter of the step size and the normal distribution of random numbers with mean 0 and variance 1; f g and f w are the current best and worst overall suitability values ​​respectively; K is a random number between 1 and -1; f i is the current value of the sparrow; f i =f g It means that the sparrows in the middle of the population are aware of the danger; if f i >f g , then the individual is at the edge of the population.

10. The method for predicting and optimizing the timing of rural domestic waste removal according to claim 1, characterized in that: The process of determining a garbage removal plan based on the amount of garbage generated in the future period includes: The maximum value of garbage accumulation is constructed. When the sum of the garbage generation in the future time period output by the model does not reach the maximum value of garbage accumulation, garbage removal is not performed; when the sum of the garbage generation in the future time period output by the model reaches the maximum value of garbage accumulation, garbage removal is performed.

Images