A Method and System for Optimizing the Efficiency of Waste Sorting Based on Data Prediction

Abstract

The present invention relates to the technical field of garbage classification prediction, and particularly relates to a method and system for optimizing garbage classification efficiency based on data prediction. The present invention obtains a load imbalance index based on the difference between the in-vehicle load and the total load at each delivery point, as well as the difference between the volume and the total volume, and then screens out the unbalanced routes; obtains an imbalance state value based on the different numbers of delivery points and load volume data on the route; and then classifies the routes; combines the load and volume of each delivery point to obtain the degree of imbalance contribution, and then screens out the delivery points with outstanding contributions; obtains a concentrated increment coefficient for the change in the amount of garbage in such delivery points; and then obtains a time interval, and predicts the amount of garbage at each delivery point with outstanding contributions in different time intervals. The present invention can accurately predict the amount of garbage at each delivery point with outstanding contributions in different time intervals.

Description

Technical Field

[0001] The present invention relates to the technical field of garbage classification prediction, and particularly to a method and system for optimizing garbage classification efficiency based on data prediction. Background Art

[0002] Problems such as resource waste, load or volume imbalance exist in the garbage collection process. Traditional methods are difficult to accurately match the changes in collection requirements caused by the fluctuations in garbage volume and density differences. Through data prediction, the distribution law and growth trend of garbage volume can be identified in advance, and the collection routes, frequencies, and vehicle types can be dynamically adjusted to ensure that the collection process is close to the full-load state, reducing the empty driving rate and resource waste. At the same time, based on the data imbalance contribution analysis and centralized increment coefficient evaluation, the high-imbalance delivery points can be accurately located, the resource allocation can be optimized, the collection efficiency can be improved, and the operation cost can be reduced.

[0003] In actual situations, it often occurs that due to the density differences among different types of garbage, the load or volume of the collection vehicle is not fully utilized during the collection process, affecting the overall collection resource allocation and scheduling optimization, and thus it is impossible to accurately predict the garbage volume at each prominent contribution delivery point in different time intervals. Summary of the Invention

[0004] In order to solve the technical problem that due to the density differences among different types of garbage, the load or volume of the collection vehicle is not fully utilized during the collection process, affecting the overall collection resource allocation and scheduling optimization, and thus it is impossible to accurately predict the garbage volume at each prominent contribution delivery point in different time intervals, the purpose of the present invention is to provide a method and system for optimizing garbage classification efficiency based on data prediction, and the specific technical solutions adopted are as follows:

[0005] A method for optimizing garbage classification efficiency based on data prediction, the method comprising:

[0006] Obtaining the garbage load data and vehicle content volume data of each garbage delivery point during each complete collection process of the collection vehicle in different preset periods;

[0007] Select a complete waste removal process within an optional preset period as the reference waste removal process; obtain the load imbalance index of the waste removal vehicle during the reference waste removal process based on the difference between all the waste load data of the waste removal vehicle and the rated total load of the waste removal vehicle during the reference waste removal process, and the difference between all the vehicle volume data and the total volume of the waste removal vehicle; screen the complete waste removal process according to the load imbalance index to obtain all the load imbalance routes; obtain the imbalance state value of each load imbalance route based on the number of waste disposal points within each load imbalance route and the waste load data and vehicle volume data of each waste disposal point; obtain the type of load imbalance to which each load imbalance route belongs based on the magnitude difference of the imbalance state values; obtain the degree of imbalance contribution of each waste disposal point in each load imbalance route based on the type of load imbalance to which each load imbalance route belongs and the waste load data and vehicle volume data of each waste disposal point; screen out the prominent contribution disposal points in each load imbalance route according to the degree of imbalance contribution;

[0008] Obtain the concentration increment coefficient of each prominent contribution disposal point each time the waste removal vehicle passes by based on the change in the amount of waste at each prominent contribution disposal point each time the waste removal vehicle passes by; cluster the concentration increment coefficients of each prominent contribution disposal point each time the waste removal vehicle passes by within each preset period to obtain all the time intervals within each preset period; predict the amount of waste at each prominent contribution disposal point in different time intervals based on the distribution characteristics of the waste load data and vehicle volume data of different prominent contribution disposal points in different time intervals.

[0009] Furthermore, the method for obtaining the load imbalance index includes:

[0010] Obtain the load imbalance index according to the load imbalance index calculation formula, and the load imbalance index calculation formula is as follows:

[0011]

[0012] In the formula, A represents the load imbalance index of the waste removal vehicle during the reference waste removal process; α end represents the total waste load data of the waste removal vehicle during the reference waste removal process; α max represents the rated total load of the waste removal vehicle; β end represents the total vehicle volume data of the waste removal vehicle during the reference waste removal process; β max represents the total volume of the waste removal vehicle; max() represents the maximum value function; || represents the absolute value function; [] represents the Iverson bracket, when the condition inside the bracket holds, the value inside the bracket is 1, when the condition inside the bracket does not hold, the value inside the bracket is 0; ∨ represents the logical or symbol.

[0013] Further, the method for obtaining the load imbalance route includes:

[0014] Taking the complete waste collection process with a load imbalance index greater than a preset first threshold as the load imbalance route of the waste collection vehicle.

[0015] Further, the method for obtaining the imbalance state value includes:

[0016] Taking the difference between the waste load data of the waste collection vehicle at the current waste disposal point and the waste load data of the previous waste disposal point as the waste load increment at the current waste disposal point;

[0017] Taking the difference between the vehicle volume data of the waste collection vehicle at the current waste disposal point and the vehicle volume data of the previous waste disposal point as the vehicle volume increment at the current waste disposal point;

[0018] Obtaining the imbalance state value according to the imbalance state value calculation formula, and the imbalance state value calculation formula is as follows:

[0019]

[0020] In the formula, B represents the imbalance state value of each load imbalance route; F represents the number of waste disposal points in each load imbalance route; α f represents the waste load increment at the f-th waste disposal point in each load imbalance route; α max represents the rated total load of the waste collection vehicle; β f represents the vehicle volume increment at the f-th waste disposal point in each load imbalance route; β max represents the total volume of the waste collection vehicle.

[0021] Further, the method for obtaining the type of load imbalance to which each load imbalance route belongs includes:

[0022] Clustering all load imbalance routes according to the imbalance state value to obtain 2 clustering clusters; the clustering clusters are divided into a volume-dominated imbalance cluster and a load-dominated imbalance cluster;

[0023] The type of load imbalance to which the load imbalance route belongs is divided into volume-dominated imbalance and load-dominated imbalance.

[0024] Further, the method for obtaining the degree of imbalance contribution includes:

[0025] In each load imbalance route, taking the waste disposal point number as the horizontal axis and the waste load data and vehicle volume data as the vertical axis, respectively establishing a waste load coordinate system and a vehicle volume coordinate system, and respectively obtaining the waste load curve and vehicle volume curve corresponding to each load imbalance route;

[0026] According to the types of load imbalance, the curve corresponding to each load imbalance route is used as the main curve, and the curves in the garbage load curve and the vehicle volume curve other than the curves corresponding to the load imbalance routes are used as the secondary curves;

[0027] Obtain the increment value corresponding to each garbage disposal point in each curve;

[0028] Obtain the degree of imbalance contribution according to the calculation formula of the degree of imbalance contribution. The calculation formula of the degree of imbalance contribution is as follows:

[0029]

[0030] In the formula, C f represents the degree of imbalance contribution of the f-th garbage disposal point in each load imbalance route; γ f represents the increment value of the main curve at the f-th garbage disposal point in each load imbalance route; δ f represents the increment value of the secondary curve at the f-th garbage disposal point in each load imbalance route; γ f-1 represents the increment value of the main curve at the (f - 1)-th garbage disposal point in each load imbalance route; H f represents the distance between the f-th garbage disposal point and the previous garbage disposal point in the load imbalance route; γ f+1 represents the increment value of the main curve at the (f + 1)-th garbage disposal point in each load imbalance route; H′ f represents the distance between the f-th garbage disposal point and the next garbage disposal point in the load imbalance route; [] represents the Iverson bracket. When the condition inside the bracket holds, the value inside the bracket is 1, and when the condition inside the bracket does not hold, the value inside the bracket is 0.

[0031] Further, the method for obtaining the prominent contribution disposal points includes:

[0032] Sort the garbage disposal points in each load imbalance route from largest to smallest according to the degree of imbalance contribution to obtain a sequence of disposal points;

[0033] In the sequence of disposal points, starting from the degree of imbalance contribution of the first garbage disposal point, add them up in turn until the sum is greater than the preset second threshold for the first time. All the garbage disposal points participating in the addition are used as the prominent contribution disposal points in the load imbalance route.

[0034] Further, the method for obtaining the concentrated increment coefficient includes:

[0035] Optionally select a prominent contribution disposal point as the reference disposal point;

[0036] Obtain the number of times the garbage truck passes through the reference disposal point and the garbage increment each time it passes through the reference disposal point in different preset periods;

[0037] Take the ratio between the garbage increment when the garbage truck passes by the reference delivery point each time and the time interval between two adjacent passes by the reference delivery point as the centralized increment coefficient at the corresponding moment when the garbage truck passes by the reference delivery point each time.

[0038] Further, according to the distribution characteristics of the garbage load data and the vehicle volume data of different prominent contribution delivery points in different time intervals, predict the garbage volume of each prominent contribution delivery point in different time intervals, including:

[0039] Take the corresponding moment when the garbage truck passes by the reference delivery point each time as the horizontal axis, and the corresponding centralized increment coefficient as the vertical axis to establish a coordinate system. Scatter all coordinate points within all preset periods in the coordinate system to obtain the centralized increment coefficient distribution map of the reference delivery point;

[0040] Cluster all coordinate points in the coordinate system to obtain all clusters. Each cluster has a maximum value and a minimum value at the corresponding moment. Projecting onto the horizontal axis where time is located is the leftmost and rightmost of the cluster. Both the leftmost and rightmost are time interpolation points. Project the time interval corresponding to each cluster onto the horizontal axis, and all time interpolation points can be obtained on the horizontal axis; after obtaining the time interpolation points, the interval between every two time difference points is a time interval;

[0041] Use the STL decomposition algorithm to obtain the reference type garbage volume trend term of the reference delivery point in each time interval of each preset period as the reference trend term; calculate the standard deviation of the reference type garbage volume of the reference delivery point in this time interval as the reference standard deviation; normalize the product between the reference trend term and the reference standard deviation to obtain the window expansion coefficient of this time interval;

[0042] Round up the product between the preset initial window length and the window expansion coefficient to obtain the time window length of this time interval;

[0043] Use the moving average method to predict the reference type garbage volume in the corresponding time interval of each next preset period;

[0044] Traverse all time intervals of all prominent contribution delivery points to obtain the prediction results of the reference type garbage volume of each prominent contribution delivery point in each time interval of each next preset period.

[0045] A garbage classification efficiency optimization system based on data prediction, the system includes a memory, a processor, and a computer program stored in the memory and operable on the processor. When the processor executes the computer program, it implements the steps of the above-mentioned garbage classification efficiency optimization method based on data prediction.

[0046] The present invention has the following beneficial effects:

[0047] Due to problems such as garbage load imbalance or vehicle volume imbalance during the garbage collection process, and there are significant differences in the amount of garbage at the same garbage disposal point during weekdays, weekends, and holidays, the present invention obtains the garbage load data and vehicle volume data of each garbage disposal point during each complete garbage collection process of the garbage truck within different preset periods; since there is a situation where the vehicle volume and the load inside the vehicle are not fully utilized during the garbage collection process, the load imbalance index of the garbage truck during the reference garbage collection process is obtained; since there are significant differences in the density of different types of garbage, all the complete garbage collection processes of the garbage truck within each preset period are screened to select the garbage collection processes that cannot achieve load balance for the garbage truck and conduct subsequent analysis on them; since different types of load imbalance result in different imbalance states of the garbage truck, the imbalance state value in each load imbalance route is calculated to analyze which situation of the type of load imbalance that is likely to be triggered by this route; in order to find out the garbage disposal points that significantly affect the complete garbage collection process and make the complete garbage collection process close to the garbage disposal points belonging to the type of load imbalance, the garbage disposal points in the complete garbage collection processes belonging to the type of load imbalance are screened to obtain the prominent contribution disposal points in each load imbalance route; since there are large differences in the garbage increment in different time intervals, first, according to the change in the garbage load data of each prominent contribution disposal point every time the garbage truck passes by, the concentrated increment coefficient of each prominent contribution disposal point every time the garbage truck passes by is obtained, and all time nodes are divided into different time intervals according to the concentrated increment coefficient; and the amount of garbage at each prominent contribution disposal point in different time intervals is predicted. The present invention takes into account the situation where the load or volume of the garbage truck is not fully utilized during the garbage collection process and can accurately predict the amount of garbage at each prominent contribution disposal point in different time intervals. Description of the Drawings

[0048] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0049] Figure 1 It is a flowchart of a method for optimizing the garbage classification efficiency based on data prediction provided by an embodiment of the present invention. Detailed Embodiments

[0050] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following specifically describes, with reference to the accompanying drawings and preferred embodiments, a method and system for optimizing garbage classification efficiency based on data prediction proposed according to the present invention, including its specific implementation manner, structure, features and effects, as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0052] The following specifically describes the specific solution of a method and system for optimizing garbage classification efficiency based on data prediction provided by the present invention with reference to the accompanying drawings.

[0053] Please refer to Figure 1 , which shows a method for optimizing garbage classification efficiency based on data prediction provided by an embodiment of the present invention. The method includes:

[0054] Step S1: Obtain the garbage load data and vehicle interior volume data of each garbage disposal point during each complete garbage collection process of the garbage collection vehicle in different preset periods.

[0055] The embodiment of the present invention is mainly applied to the garbage volume prediction scenario of garbage disposal points. Due to problems such as garbage load imbalance or vehicle interior volume imbalance during the garbage collection process, the embodiment of the present invention obtains the garbage load data and vehicle interior volume data of each garbage disposal point during each complete garbage collection process of the garbage collection vehicle. Since the amount of garbage at the same garbage disposal point varies greatly during weekdays, weekends and holidays, the garbage load data and vehicle interior volume data of each garbage disposal point in different preset periods are statistically analyzed and subsequent time intervals are divided.

[0056] In one embodiment of the present invention, the preset period is set to 1 day. It should be noted that the preset period can be set by oneself and is not limited here.

[0057] In the embodiment of the present invention, the description is mainly based on weekdays. Since weekends and holidays only differ in category and do not affect the analysis process, the analysis processes for weekends and holidays are the same as those for weekdays.

[0058] Step S2: Select a complete garbage removal process within a preset period as the reference garbage removal process; obtain the load imbalance index of the garbage truck during the reference garbage removal process according to the difference between all the garbage load data of the garbage truck during the reference garbage removal process and the rated total load of the garbage truck, and the difference between all the vehicle interior volume data and the total volume of the garbage truck; screen the complete garbage removal process according to the load imbalance index to obtain all the load imbalance routes; obtain the imbalance state value of each load imbalance route according to the number of garbage disposal points within each load imbalance route, and the garbage load data and vehicle interior volume data of each garbage disposal point; obtain the type of load imbalance to which each load imbalance route belongs according to the magnitude difference of the imbalance state values; obtain the imbalance contribution degree of each garbage disposal point in each load imbalance route according to the type of load imbalance to which each load imbalance route belongs, and the garbage load data and vehicle interior volume data of each garbage disposal point; and screen out the prominent contribution disposal points in each load imbalance route according to the imbalance contribution degree.

[0059] For each garbage truck, there is a maximum load capacity that can be carried and a maximum volume that can be accommodated during a garbage removal process. When the vehicle is defined as load imbalance, it means the following situation: that is, after completing a garbage removal operation, the difference between the cumulative garbage load data of the garbage truck and the rated total load of the garbage truck is large, and the difference between the cumulative vehicle interior volume data of the vehicle and the total volume of the garbage truck is large. The occurrence of this situation indicates that the garbage truck is not fully utilized during the current garbage removal process. Therefore, in the embodiments of the present invention, the load imbalance index of the garbage truck during the reference garbage removal process is obtained according to the difference between all the garbage load data of the garbage truck during the reference garbage removal process and the rated total load of the garbage truck, and the difference between all the vehicle interior volume data and the total volume of the garbage truck.

[0060] Preferably, in an embodiment of the present invention, the method for obtaining the load imbalance index includes:

[0061] Obtain the load imbalance index according to the load imbalance index calculation formula, and the load imbalance index calculation formula is as follows:

[0062]

[0063] In the formula, A represents the load imbalance index of the garbage truck during the reference garbage removal process; α end represents the total garbage load data of the garbage truck during the reference garbage removal process; α max represents the rated total load of the garbage truck; β end represents the total vehicle interior volume data of the garbage truck during the reference garbage removal process; β maxrepresents the total volume of the garbage truck; max() represents the maximum value function; || represents the absolute value function; [] represents the Iverson bracket, when the condition inside the bracket holds, the value inside the bracket is 1, and when the condition inside the bracket does not hold, the value inside the bracket is 0; ∨ represents the logical OR symbol.

[0064] In the calculation formula of the load imbalance index, when the condition inside the Iverson bracket holds, it indicates that the garbage truck does not end in a fully loaded state during the reference garbage collection process. At this time, when [(α end -α max ) ≤ 0 ∨ β end > 0], the value is 1, and at this time, the load imbalance index of the garbage truck during the reference garbage collection process is calculated; and when or is larger, it indicates that the load imbalance state of the garbage truck after the reference garbage collection process is more serious, and at this time, the load imbalance index of the garbage truck during the reference garbage collection process is larger.

[0065] The occurrence of load imbalance in the garbage truck during the garbage collection process is due to the load contradiction caused by the physical property differences of different classified garbage received. Specifically, the densities of different types of garbage are significantly different. Recyclable garbage, such as low-density garbage like plastics and paper, is easy to reach the volume limit, while wet garbage has the characteristic of high density and is therefore more likely to reach the load limit. All complete garbage collection processes of the garbage truck within each preset period are screened to select the garbage collection processes that cannot make the garbage truck reach load balance, and subsequent analysis is carried out on them.

[0066] Preferably, in an embodiment of the present invention, the method for obtaining the load imbalance route includes:

[0067] Taking the complete garbage collection process with a load imbalance index greater than a preset first threshold as the load imbalance route of the garbage truck. In an embodiment of the present invention, the preset first threshold is set to 0.8. It should be noted that the preset first threshold can be set by oneself and is not limited here.

[0068] In reality, the load imbalance of the garbage truck is related to the type of garbage being collected, that is, high-density garbage, such as wet garbage, is easy to trigger load-dominated imbalance, while low-density garbage, such as dry garbage, is easy to trigger volume-dominated imbalance. Therefore, in the embodiment of the present invention, the imbalance state value in each load imbalance route is calculated to analyze which of the two imbalances the route is prone to trigger.

[0069] Preferably, in an embodiment of the present invention, the method for obtaining the imbalance state value includes:

[0070] Taking the difference between the garbage load data of the garbage truck at the current garbage disposal point and the garbage load data at the previous garbage disposal point as the garbage load increment at the current garbage disposal point.

[0071] Take the difference between the vehicle volume data at the current waste disposal point and the vehicle volume data at the previous waste disposal point of the waste collection vehicle as the vehicle volume increment at the current waste disposal point.

[0072] Obtain the imbalance state value according to the imbalance state value calculation formula, and the imbalance state value calculation formula is as follows:

[0073]

[0074] In the formula, B represents the imbalance state value of each load imbalance route; F represents the number of waste disposal points on each load imbalance route; α f represents the waste load increment at the f-th waste disposal point on each load imbalance route; α max represents the rated total load of the waste collection vehicle; β f represents the vehicle volume increment at the f-th waste disposal point on each load imbalance route; β max represents the total volume of the waste collection vehicle.

[0075] In the imbalance state value calculation formula, represents the waste load growth rate at the f-th waste disposal point on each load imbalance route, represents the vehicle volume growth rate at the f-th waste disposal point on each load imbalance route; at the f-th waste disposal point, if the waste load growth rate is greater than the vehicle volume growth rate, it indicates that the f-th waste disposal point is more inclined to be load-dominated imbalance, and if the waste load growth rate is less than the vehicle volume growth rate, it indicates that the f-th waste disposal point is more inclined to be volume-dominated imbalance; analyze all waste disposal points to obtain the imbalance state value of each load imbalance route.

[0076] Preferably, in the embodiment of the present invention, the method for obtaining the type of load imbalance to which each load imbalance route belongs includes:

[0077] The imbalance state value reflects which dominant imbalance state the corresponding waste collection route tends to be in the current waste collection process. The larger the imbalance state value, the more inclined it is to be load-dominated imbalance, and the smaller the imbalance state value, the more inclined it is to be volume-dominated imbalance. Therefore, the K-means clustering algorithm is used here, with the number of clustering clusters set to 2. After clustering the imbalance state values, two clustering clusters are obtained as the types of load imbalance to which the load imbalance routes belong. The type of load imbalance corresponding to the imbalance state value in the clustering cluster with the smaller mean belongs to volume-dominated imbalance, while the type of load imbalance corresponding to the imbalance state value in the other cluster belongs to load-dominated imbalance.

[0078] Next, the garbage disposal points in the complete garbage removal process with the corresponding load imbalance types are screened. The purpose is to find out the garbage disposal points that significantly affect the complete garbage removal process and make the complete garbage removal process approach the load imbalance types. That is, the reason for the load imbalance or volume imbalance in the complete garbage removal process is that the amount of garbage of the corresponding type in some garbage disposal points is excessive. Therefore, in the embodiments of the present invention, according to the load imbalance types to which each load imbalance route belongs, and the garbage load data and vehicle volume data of each garbage disposal point, the imbalance contribution degree of each garbage disposal point in each load imbalance route is obtained; and the prominent contribution disposal points in each load imbalance route are screened according to the imbalance contribution degree.

[0079] Preferably, in the embodiments of the present invention, the method for obtaining the imbalance contribution degree includes:

[0080] In each load imbalance route, taking the garbage disposal point number as the horizontal axis and the garbage load data and vehicle volume data as the vertical axis, a garbage load coordinate system and a vehicle volume coordinate system are respectively established, and the garbage load curve and the vehicle volume curve corresponding to each load imbalance route are respectively obtained, and the starting point and the ending point of the garbage load curve and the vehicle volume curve are the same.

[0081] Since there are two types of garbage in any load imbalance route, it is necessary to analyze the changes of the garbage load curve and the vehicle volume curve at the same time. According to the load imbalance types, the curve corresponding to each load imbalance route is used as the main curve, and the other curve is used as the secondary curve; that is, for the load imbalance route with load-dominated imbalance, the garbage load curve is used as the main curve and the vehicle volume curve is used as the secondary curve, and for the load imbalance route with volume-dominated imbalance, the vehicle volume curve is used as the main curve and the garbage load curve is used as the secondary curve.

[0082] Calculate the difference between all the garbage of the garbage type corresponding to each curve of each garbage disposal point and all the garbage of the garbage type corresponding to each curve of the previous garbage disposal point, and use it as the incremental value corresponding to each garbage disposal point in each curve.

[0083] The imbalance contribution degree is obtained according to the imbalance contribution degree calculation formula. The imbalance contribution degree calculation formula is as follows:

[0084]

[0085] In the formula, C f represents the imbalance contribution degree of the f-th garbage disposal point in each load imbalance route; γ f represents the incremental value of the main curve at the f-th garbage disposal point in each load imbalance route; δ f represents the incremental value of the secondary curve at the f-th garbage disposal point in each load imbalance route; γf-1 Denote the incremental value of the main curve at the (f - 1)-th waste disposal point in each load imbalance route; H f Denote the distance between the f-th and the previous waste disposal points in the load imbalance route; γ f+1 Denote the incremental value of the main curve at the (f + 1)-th waste disposal point in each load imbalance route; H′ f Denote the distance between the f-th and the next waste disposal points in the load imbalance route; [] represents the Iverson bracket, whose value is 1 when the condition inside the bracket holds, and 0 when the condition does not hold.

[0086] In the calculation formula of the imbalance contribution degree, when the condition inside the Iverson bracket holds, the incremental value of the main curve is greater than that of the secondary curve. At this time, calculate the imbalance contribution degree of each waste disposal point in each load imbalance route; and the difference γ between the incremental value of the main curve and that of the secondary curve at the f-th waste disposal point f -δ f The larger it is, the more the f-th waste disposal point tends to collect the waste type corresponding to the main curve. At this time, the imbalance contribution degree of the f-th waste disposal point is greater; in addition, the generation and classification of waste types have significant geographical aggregation. For example, commercial areas mostly generate wet waste, and areas closer to commercial areas are more likely to generate more wet waste due to the radiation of commercial areas. Then, load imbalance is more likely to be triggered during the cleaning work in this area. Therefore, (γ f-1 +H f )+(γ f+1 +H′ f ) The larger it is, the more likely this route is located in an area close to the commercial area and is affected by the radiation of the commercial area, and is prone to generate wet waste, that is, it is more likely to be affected by load imbalance. At this time, the area near the f-th waste disposal point is more likely to have waste types corresponding to the main curve.

[0087] Preferably, in the embodiments of the present invention, the method for obtaining the prominent contribution disposal points includes:

[0088] Sort the waste disposal points in each load imbalance route in descending order of the imbalance contribution degree to obtain a sequence of disposal points.

[0089] In the sequence of disposal points, add up the imbalance contribution degrees of the waste disposal points starting from the first one until the sum is greater than the preset second threshold for the first time. Then, all the waste disposal points participating in the addition are used as the prominent contribution disposal points in the load imbalance route. In an embodiment of the present invention, the preset second threshold is set to 50%. It should be noted that the preset second threshold can be set by oneself and is not limited herein.

[0090] Step S3: According to the change in the amount of garbage at each prominent contribution delivery point each time the garbage truck passes by, obtain the centralized increment coefficient of each prominent contribution delivery point each time the garbage truck passes by; cluster the centralized increment coefficients of each prominent contribution delivery point each time the garbage truck passes by within each preset period to obtain all time intervals within each preset period; according to the distribution characteristics of the garbage load data and the vehicle volume data of different prominent contribution delivery points within different time intervals, predict the amount of garbage at each prominent contribution delivery point in different time intervals.

[0091] According to the above steps, prominent contribution delivery points can be obtained, and then the amount of garbage at the prominent contribution delivery points can be predicted. The object of prediction is the amount of garbage corresponding to the main curve type among the prominent contribution delivery points. In the embodiment of the present invention, the garbage generation time is divided into weekdays, weekends and holidays, and only the garbage types corresponding to the main curve are discussed. Therefore, long-term trends and seasonal factors are avoided. In the embodiment of the present invention, the moving average method is used to predict the amount of garbage at each prominent contribution delivery point in different time intervals, which can effectively smooth the noise, reflect the basic fluctuation law, and is suitable for short-term prediction to support real-time scheduling optimization.

[0092] Since there are large differences in the garbage increments in different time intervals, it is necessary to make the time window length required by the moving average method adaptive. Therefore, in the embodiment of the present invention, first, according to the change in the garbage load data of each prominent contribution delivery point each time the garbage truck passes by, obtain the centralized increment coefficient of each prominent contribution delivery point each time the garbage truck passes by, and divide all time nodes into three time intervals: weekdays, weekends and holidays according to the centralized increment coefficient; and predict the amount of garbage at each prominent contribution delivery point in different time intervals.

[0093] Preferably, in an embodiment of the present invention, the method for obtaining the centralized increment coefficient includes:

[0094] Optionally select a prominent contribution delivery point as a reference delivery point; obtain the number of times the garbage truck passes by the reference delivery point within different preset periods and the garbage increment each time the garbage truck passes by the reference delivery point.

[0095] Take the garbage type corresponding to the main curve when the garbage truck passes by the reference delivery point as the reference type;

[0096] Take the ratio between the garbage increment each time the garbage truck passes by the reference delivery point and the time interval between every two adjacent times the garbage truck passes by the reference delivery point as the centralized increment coefficient of the reference delivery point at the corresponding moment each time the garbage truck passes by. In the embodiment of the present invention, the calculation formula of the centralized increment coefficient is as follows:

[0097]

[0098] Wherein, D represents the centralized increment coefficient corresponding to the reference delivery point at the corresponding moment when the garbage truck passes each time; K i represents the increment of the reference type of garbage when the garbage truck passes the reference delivery point for the i-th time; T i represents the time interval between the corresponding moment when the garbage truck passes the reference delivery point for the i-th time and the corresponding moment when it passes the reference delivery point for the (i - 1)-th time.

[0099] In the calculation formula of the centralized increment coefficient, if more increment of the reference type of garbage is generated at the reference delivery point in a relatively short time, that is, T i is smaller and K i is larger, it indicates that the centralized increment coefficient at the corresponding moment when the garbage truck passes the reference delivery point for the i-th time is larger.

[0100] Preferably, the embodiment of the present invention provides a prediction method, and the specific steps are as follows:

[0101] Taking the corresponding moment when the garbage truck passes the reference delivery point each time as the horizontal axis and the corresponding centralized increment coefficient as the vertical axis to establish a coordinate system, and spreading all the coordinate points within all preset periods in the coordinate system to obtain the distribution diagram of the centralized increment coefficient of the reference delivery point.

[0102] Using the LOF algorithm to cluster all the coordinate points in the coordinate system to obtain all the clustering clusters. Each cluster has a maximum value and a minimum value at the corresponding moment. Projecting them onto the horizontal axis where time is located is the leftmost and rightmost sides of the clustering cluster. Both the leftmost and rightmost sides are time interpolation points. Projecting the time interval corresponding to each cluster onto the horizontal axis can obtain all the time interpolation points on the horizontal axis. After obtaining the time interpolation points, the interval between every two time difference points is a time interval.

[0103] So far, the acquisition of all time intervals of the reference delivery point within each preset period is completed.

[0104] Using the STL decomposition algorithm to obtain the reference type garbage volume trend item of the reference delivery point under each time interval within each preset period as the reference trend item; calculating the standard deviation of the reference type garbage volume of the reference delivery point under this time interval as the reference standard deviation; normalizing the product between the reference trend item and the reference standard deviation to obtain the window expansion coefficient of this time interval.

[0105] Rounding up the product of the preset initial window length and the window expansion coefficient to obtain the time window length of this time interval.

[0106] Using the moving average method to predict the reference type garbage volume corresponding to the time interval within each next preset period.

[0107] Traverse all time intervals of all prominent contribution delivery points, and obtain the predicted results of the reference type of garbage volume for each time interval of each prominent contribution delivery point in each preset period in the future.

[0108] In summary, obtain the garbage load data and vehicle content volume data of each garbage delivery point during each complete garbage collection process within different preset periods; select a complete garbage collection process within a preset period as the reference garbage collection process; according to the difference between all the garbage load data of the garbage truck during the reference garbage collection process and the rated total load of the garbage truck, and the difference between all the vehicle content volume data and the total volume of the garbage truck, obtain the load imbalance index of the garbage truck during the reference garbage collection process; screen the complete garbage collection process according to the load imbalance index to obtain all the load imbalance routes; according to the number of garbage delivery points within each load imbalance route, and the garbage load data and vehicle content volume data of each garbage delivery point, obtain the imbalance state value of each load imbalance route; according to the magnitude difference of the imbalance state values, obtain the type of load imbalance to which each load imbalance route belongs; according to the type of load imbalance to which each load imbalance route belongs, and the garbage load data and vehicle content volume data of each garbage delivery point, obtain the degree of imbalance contribution of each garbage delivery point in each load imbalance route; screen out the prominent contribution delivery points in each load imbalance route according to the degree of imbalance contribution; according to the change of the garbage load data of each prominent contribution delivery point each time the garbage truck passes by, obtain the concentration increment coefficient of each prominent contribution delivery point each time the garbage truck passes by; cluster the concentration increment coefficients of each prominent contribution delivery point each time the garbage truck passes by within each preset period to obtain all the time intervals within each preset period; according to the distribution characteristics of the garbage load data and vehicle content volume data of different prominent contribution delivery points within different time intervals, predict the garbage volume of each prominent contribution delivery point within different time intervals.

[0109] An embodiment of the present invention provides a garbage classification efficiency optimization system based on data prediction. The system includes a memory, a processor, and a computer program, where the memory is used to store the corresponding computer program, the processor is used to run the corresponding computer program, and when the computer program runs in the processor, it can implement the methods described in steps S1 - S3.

[0110] It should be noted that: the above sequence of embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0111] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the differences between each embodiment and other embodiments are emphasized.

Claims

1. A method for optimizing the efficiency of waste classification based on data prediction, characterized in that, The method includes: Obtaining the garbage load data and vehicle volume data of each garbage disposal point during each complete garbage collection process of the garbage truck in different preset periods; Optionally selecting a complete garbage collection process within a preset period as a reference garbage collection process; obtaining the load imbalance index of the garbage truck during the reference garbage collection process according to the difference between all the garbage load data of the garbage truck during the reference garbage collection process and the rated total load of the garbage truck, and the difference between all the vehicle volume data and the total volume of the garbage truck; screening the complete garbage collection process according to the load imbalance index to obtain all the load imbalance routes; obtaining the imbalance state value of each load imbalance route according to the number of garbage disposal points within each load imbalance route, and the garbage load data and vehicle volume data of each garbage disposal point; obtaining the type of load imbalance to which each load imbalance route belongs according to the magnitude difference of the imbalance state value; obtaining the contribution degree of imbalance of each garbage disposal point in each load imbalance route according to the type of load imbalance to which each load imbalance route belongs, and the garbage load data and vehicle volume data of each garbage disposal point; screening out the prominent contribution disposal points in each load imbalance route according to the contribution degree of imbalance; Obtaining the concentration increment coefficient of each prominent contribution disposal point each time the garbage truck passes according to the change in the amount of garbage at each prominent contribution disposal point each time the garbage truck passes; clustering the concentration increment coefficients of each prominent contribution disposal point each time the garbage truck passes within each preset period to obtain all time intervals within each preset period; predicting the amount of garbage at each prominent contribution disposal point in different time intervals according to the distribution characteristics of the garbage load data and vehicle volume data of different prominent contribution disposal points in different time intervals; The method for obtaining the contribution degree of imbalance includes: In each load imbalance route, a garbage load coordinate system and a vehicle volume coordinate system are respectively established with the garbage disposal point serial number as the horizontal axis and the garbage load data and vehicle volume data as the vertical axis, and the garbage load curve and vehicle volume curve corresponding to each load imbalance route are respectively obtained; According to the type of load imbalance, one of the garbage load curve and vehicle volume curve corresponding to each load imbalance route is used as the main curve, and the other curve is used as the sub - curve; Obtaining the increment value corresponding to each garbage disposal point in each curve; Obtaining the contribution degree of imbalance according to the contribution degree of imbalance calculation formula, and the contribution degree of imbalance calculation formula is as follows: where C f represents the imbalance contribution degree of the f-th garbage disposal point in each load imbalance route; γ f represents the incremental value of the main curve at the f-th garbage disposal point in each load imbalance route; δ f represents the incremental value of the secondary curve at the f-th garbage disposal point in each load imbalance route; γ f-1 represents the incremental value of the main curve at the (f - 1)-th garbage disposal point in each load imbalance route; H f represents the distance between the f-th garbage disposal point and the previous garbage disposal point in the load imbalance route; γ f+1 represents the incremental value of the main curve at the (f + 1)-th garbage disposal point in each load imbalance route; H′ f represents the distance between the f-th garbage disposal point and the next garbage disposal point in the load imbalance route; [] represents the Iverson bracket, whose value is 1 when the condition inside the bracket holds and 0 when the condition inside the bracket does not hold.

2. The method for optimizing the garbage classification efficiency based on data prediction according to claim 1, wherein The method for obtaining the load imbalance index includes: Obtaining the load imbalance index according to the load imbalance index calculation formula, and the load imbalance index calculation formula is as follows: Wherein, A represents the load imbalance index of the garbage truck in the reference garbage collection process; α end represents the total garbage load data of the garbage truck in the reference garbage collection process; α max represents the rated total load of the garbage truck; β end represents the total in-vehicle volume data of the garbage truck in the reference garbage collection process; β max represents the total volume of the garbage truck; max() represents the maximum value function; || represents the absolute value function; [] represents the Iverson bracket, when the condition inside the bracket holds, the value inside the bracket is 1, when the condition inside the bracket does not hold, the value inside the bracket is 0; ∨ represents the logical or symbol.

3. The method for optimizing the garbage classification efficiency based on data prediction according to claim 1, wherein The method for obtaining the load imbalance route includes: Regarding the complete garbage collection process with the load imbalance index greater than the preset first threshold as the load imbalance route of the garbage truck.

4. A method for optimizing the garbage classification efficiency based on data prediction according to claim 1, characterized in that, The method for obtaining the imbalance state value includes: Taking the difference between the garbage load data of the garbage truck at the current garbage disposal point and the garbage load data of the previous garbage disposal point as the garbage load increment of the current garbage disposal point; Take the difference between the vehicle volume data of the garbage collection vehicle at the current garbage disposal point and the vehicle volume data at the previous garbage disposal point as the vehicle volume increment at the current garbage disposal point. Obtain the imbalance state value according to the imbalance state value calculation formula, and the imbalance state value calculation formula is as follows: Wherein, B represents the imbalance state value of each load imbalance route; F represents the number of garbage disposal points on each load imbalance route; α f represents the garbage load increment of the f-th garbage disposal point on each load imbalance route; α max represents the rated total load of the garbage truck; β f represents the vehicle content volume increment of the f-th garbage disposal point on each load imbalance route; β max represents the total volume of the garbage truck.

5. A method for optimizing the garbage classification efficiency based on data prediction according to claim 1, characterized in that, The method for obtaining the type of load imbalance to which each load imbalance route belongs includes: Cluster all load imbalance routes according to the imbalance state value to obtain 2 clusters; the clusters are divided into a volume-dominated imbalance cluster and a load-dominated imbalance cluster. The types of load imbalance to which the load imbalance routes belong are divided into volume-dominated imbalance and load-dominated imbalance.

6. The method for optimizing the garbage classification efficiency based on data prediction according to claim 1, characterized in that The method for obtaining the prominent contribution disposal points includes: Sort the garbage disposal points in each load imbalance route in descending order according to the imbalance contribution degree to obtain a disposal point sequence. In the disposal point sequence, add up the imbalance contribution degrees of the garbage disposal points from the first one in turn until the sum is greater than the preset second threshold for the first time. Then, take all the garbage disposal points participating in the addition as the prominent contribution disposal points in the load imbalance route.

7. A method for optimizing the garbage classification efficiency based on data prediction according to claim 1, characterized in that The method for obtaining the concentrated increment coefficient includes: Arbitrarily select a prominent contribution disposal point as the reference disposal point. Obtain the number of times the garbage collection vehicle passes through the reference disposal point and the garbage increment each time it passes through the reference disposal point in different preset periods. Take the ratio of the garbage increment each time the garbage collection vehicle passes through the reference disposal point to the time interval between every two adjacent passes through the reference disposal point as the concentrated increment coefficient at the corresponding moment each time the garbage collection vehicle passes through the reference disposal point.

8. A method for optimizing the garbage classification efficiency based on data prediction according to claim 1, characterized in that, Predict the garbage volume of each prominent contribution disposal point in different time intervals according to the distribution characteristics of the garbage load data and vehicle volume data of different prominent contribution disposal points in different time intervals, including: Take the corresponding moment each time the garbage collection vehicle passes through the reference disposal point as the horizontal axis and the corresponding concentrated increment coefficient as the vertical axis to establish a coordinate system. Scatter all the coordinate points in all preset periods in the coordinate system to obtain the concentrated increment coefficient distribution map of the reference disposal point. Cluster all the coordinate points in the coordinate system to obtain all clusters. Each cluster has a maximum value and a minimum value at the corresponding moment. Projecting them onto the horizontal axis where time is located is the leftmost and rightmost sides of the cluster. Both the leftmost and rightmost sides are time interpolation points. Project the time intervals corresponding to each cluster onto the horizontal axis, and all the time interpolation points can be obtained on the horizontal axis. After obtaining the time interpolation points, the interval between every two time difference points is a time interval. Use the STL decomposition algorithm to obtain the reference type garbage volume trend term of the reference disposal point in each time interval of each preset period as the reference trend term; calculate the standard deviation of the reference type garbage volume of the reference disposal point in this time interval as the reference standard deviation; normalize the product of the reference trend term and the reference standard deviation to obtain the window expansion coefficient of this time interval. Round up the product of the preset initial window length and the window expansion coefficient to obtain the time window length of this time interval. Use the moving average method to predict the reference type garbage volume in the corresponding time interval in each subsequent preset period. Traverse all time intervals of all prominent contribution placement points, and obtain the predicted results of the reference type of garbage volume for each time interval of each prominent contribution placement point in each preset period in the next.

9. A garbage classification efficiency optimization system based on data prediction, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of a garbage classification efficiency optimization method based on data prediction according to any one of claims 1 to 8.

Images