A smart sanitation operation planning system and method

Abstract

This invention relates to the field of sanitation operation planning technology, and discloses a smart sanitation operation planning system and method. The smart sanitation operation planning system includes: a data acquisition module for collecting regional feature data of M sub-regions of the area to be cleaned and vehicle feature data of N sanitation vehicles; a data preprocessing module for preprocessing the regional feature data and vehicle feature data; an operation time prediction model construction module for constructing an operation time prediction model; a planning scheme generation module for generating a planning scheme based on the operation time prediction model and a Cauchy mutation strategy-incorporated hog optimization algorithm; and allocating the M sub-regions to the N sanitation vehicles according to the planning scheme. This invention achieves collaborative operation of multiple sanitation vehicles by constructing an operation time prediction model combined with a Cauchy mutation strategy-incorporated hog optimization algorithm, improving the efficiency of sanitation operations and realizing efficient resource allocation.

Description

Technical Field

[0001] The present invention relates to the technical field of sanitation operation planning, and more specifically, to an intelligent sanitation operation planning system and method. Background Art

[0002] With the popularization of technologies such as the Internet of Things, big data, and cloud computing, the demand for sanitation operations has shifted from simple cleaning to a comprehensive pursuit of efficiency, intelligence, and sustainability. In this context, smart sanitation operation planning has become particularly important. Currently, smart sanitation operation planning is mostly done by dividing the area to be cleaned into multiple sub-areas, finding a shortest path that traverses all sub-areas, and then using the A* or D* algorithm to plan the path for each sub-area, and finally obtaining a shortest path for the trash can that fully covers the cleaning area.

[0003] However, the above methods mainly focus on the operation planning of a single sanitation vehicle, and do not consider the collaborative operation of multiple vehicles, such as how to effectively assign tasks to different sanitation vehicles and how to ensure that multiple sanitation vehicles do not appear on the same road to cause traffic jams.

[0004] Therefore, there is an urgent need for a smart sanitation operation planning system and method to solve the above problems. Summary of the invention

[0005] The present invention provides an intelligent sanitation operation planning system to solve the technical problem that the related technology only considers the operation planning of a single sanitation vehicle but does not consider the coordinated operation of multiple sanitation vehicles.

[0006] The present invention provides a smart sanitation operation planning system, comprising:

[0007] A data acquisition module, which is used to collect regional characteristic data of M sub-areas of the area to be cleaned and vehicle characteristic data of N sanitation vehicles;

[0008] Regional characteristic data include: traffic flow index, number of traffic lights, number of garbage collection points, number of vehicles, total length of roads and area of ​​the region;

[0009] Vehicle characteristic data include: average travel speed, loading efficiency, unloading efficiency and maximum loading capacity;

[0010] The regional feature data of the M sub-regions are represented by the first feature sequence, and the regional feature data of the m-th sub-region is represented by the m-th sequence unit of the first feature sequence, where 1≤m≤M;

[0011] The vehicle characteristic data of N sanitation vehicles are represented by the second characteristic sequence, and the vehicle characteristic data of the nth sanitation vehicle is represented by the nth sequence unit of the second characteristic sequence, where 1≤n≤N;

[0012] A data preprocessing module, which is used to preprocess the first feature sequence and the second feature sequence to obtain a third feature sequence and a fourth feature sequence;

[0013] An operation time prediction model building module, which is used to build an operation time prediction model according to the third feature sequence and the fourth feature sequence;

[0014] The input of the job time prediction model is the third feature sequence and the fourth feature sequence, and the output is the estimated completion time;

[0015] A planning scheme generation module is used to generate a planning scheme based on the operation time prediction model combined with the crown porcupine optimization algorithm that introduces the Cauchy mutation strategy;

[0016] According to the planning scheme, M sub-areas are allocated to N sanitation vehicles.

[0017] Furthermore, M, N, m and n are all custom parameters, where M>N.

[0018] Furthermore, the formula for the traffic flow index of the mth sub-area is: ,in represents the traffic flow index of the mth sub-area, represents the number of sanitation vehicles in the mth sub-area, represents the area of ​​the mth sub-region;

[0019] The formula for the loading efficiency of the nth sanitation vehicle is: ,in represents the loading efficiency of the nth sanitation vehicle, represents the maximum loading capacity of the nth sanitation vehicle, represents the average time it takes for the nth sanitation truck to be fully loaded with garbage;

[0020] The formula for the unloading efficiency of the nth sanitation vehicle is: ,in represents the unloading efficiency of the nth sanitation vehicle, It represents the average time for the nth sanitation truck to unload a full load of garbage.

[0021] Furthermore, the step of preprocessing the first feature sequence and the second feature sequence includes:

[0022] Step S201, for missing values ​​in the first characteristic sequence and the second characteristic sequence, filling processing is performed by taking the average value of the value of the previous sequence unit and the value of the next sequence unit;

[0023] Step S202, removing duplicate values ​​in the first feature sequence and the second feature sequence;

[0024] Step S203: normalize the first feature sequence and the second feature sequence using a z-score normalization method.

[0025] Further, the operation time prediction model includes: a first hidden layer, a second hidden layer, a separation layer, a combination layer and a classifier;

[0026] The first hidden layer inputs the third feature sequence and outputs the first updated feature;

[0027] The second hidden layer includes N hidden units, the nth hidden unit inputs the nth sequence unit of the fourth feature sequence, and outputs the second hidden feature;

[0028] The separation layer is used to separate the second hidden feature output by the Nth hidden unit as the second update feature;

[0029] The combination layer is used to expand the first update feature into a vector representation and concatenate it with the second update feature to obtain a combined vector;

[0030] The input of the classifier is a combination vector, and the classification space of the classifier represents the estimated completion time.

[0031] Furthermore, the formula for the first hidden layer is:

[0032] ;

[0033] ;

[0034] Where A represents the first updated feature of the first hidden layer output, e represents the attention score, and denote the first weight parameter and the second weight parameter respectively, represents the first weight vector, Q, K and V represent the first intermediate feature, the second intermediate feature and the third intermediate feature of the first hidden layer, The number of dimensions of each sequence unit representing the third characteristic sequence is a constant value of 6, and T represents a transposition operation.

[0035] Furthermore, the second hidden layer is constructed based on gated recurrent units.

[0036] Furthermore, a simulation platform is used to simulate the area to be cleaned and the sanitation vehicle, and training data for different scenarios and label data of corresponding instances are generated based on the simulation platform, wherein the label data is represented by the time when the sanitation vehicle handles the garbage collection point in the sub-area.

[0037] Furthermore, the steps of the crested porcupine optimization algorithm using the Cauchy mutation strategy used in the planning scheme generation module include:

[0038] Step S401, generating an initialization population with P number of individuals meeting the constraint conditions, where P is a custom parameter;

[0039] The constraints include: each sanitation vehicle is assigned to at least one sub-area; each sub-area can only be assigned to one sanitation vehicle; the sub-areas assigned to each sanitation vehicle must be adjacent;

[0040] The individuals of the initialized population are represented by matrix coding, which is expressed as: ,in They represent the allocation of the first sub-area to the Mth sub-area to the first sanitation vehicle, They represent the allocation of the first sub-area to the first sanitation vehicle to the Nth sanitation vehicle, respectively. Indicates the allocation of the Mth sub-area to the Nth sanitation vehicle, and the element value of the matrix code is represented by 0 or 1;

[0041] Step S402, constructing an objective function by combining the operation time prediction model to calculate and obtain the fitness value of each individual in the initialization population;

[0042] The objective function is calculated as follows: , where Fitness represents the fitness value of the individual, It means that the regional characteristic data of the mth sub-region and the vehicle characteristic data of the nth sanitation vehicle are input into the estimated completion time output by the operation time prediction model. Represents the weight parameter of the objective function, which is a custom parameter;

[0043] Step S403, generating a random number ranging from 0 to 1, and when the random number is less than a first threshold, performing global exploration, otherwise performing local development, wherein the first threshold is a custom parameter;

[0044] Step S4041, during global exploration, a random number ranging from 0 to 1 is generated. When the random number is less than a second threshold, the first defense strategy is adopted, otherwise the second defense strategy is adopted, wherein the first threshold is a custom parameter;

[0045] Step S4042: during local development, a random number ranging from 0 to 1 is generated. When the random number is less than the second threshold, the third defense strategy is adopted, otherwise the fourth defense strategy is adopted.

[0046] The formula for the first defense strategy is:

[0047] ;

[0048] ;

[0049] The formula for the second defense strategy is:

[0050] ;

[0051] The formula for the third defense strategy is:

[0052] ;

[0053] ;

[0054] ;

[0055] ;

[0056] The formula for the fourth defense strategy is:

[0057] ;

[0058] in represents the position of the i-th individual when the number of iterations is k+1, represents the position of the i-th individual when the number of iterations is k, represents random numbers that follow a normal distribution, , , , and rand is a random number between 0 and 1, represents the global optimal solution when the number of iterations is k, represents the position of the predator when the number of iterations is k, r is a random integer between 1 and P, represents a binary vector consisting entirely of 0s, then , the predator no longer moves towards the crested porcupine, when When contains 1, the second defense strategy is based on two random individuals and To determine whether the predator is approaching or moving away from the crested porcupine, r1, r2, and r3 are random integers between 1 and P. represents the first vector parameter used to control the search direction, represents the defense factor when the number of iterations is k, It represents the odor diffusion factor when the number of iterations is k, where k is the current number of iterations. is the maximum number of iterations, represents the fitness of the i-th individual when the number of iterations is k, For the second parameter, avoid division by 0. is the speed convergence factor, exp represents the exponential function;

[0059] Step S405, mutating the individuals in the population based on the Cauchy mutation strategy;

[0060] The probability density function of the Cauchy distribution is: ,in represents the location parameter, γ represents the scale parameter, and x represents the custom variable;

[0061] Step S4051, determine the location parameter and scale parameter of the Cauchy distribution, and adjust the scale parameter γ according to the current number of iterations. ,in represents the scale parameter when the number of population iterations is k+1, represents the scale parameter when the number of population iterations is k, and β represents a decreasing factor between 0 and 1;

[0062] Step S4052, determining the variation ratio of individuals in the population;

[0063] Step S4053, generating a mutant individual, for the selected individual, randomly extracting a mutation value from the Cauchy distribution, and applying it to each element value of the individual, the specific application method is: when the mutation value is greater than the third threshold, flipping the element value mapped by the mutation value, that is, 0 becomes 1, and 1 becomes 0;

[0064] Step S406, after the current iteration is completed, the number of populations is reduced according to the cyclic population reduction technique, and the scale parameter is updated;

[0065] Step S407, when the current number of iterations is greater than or equal to the maximum number of iterations, the individual with the largest fitness in the initialized population is used as the planning scheme, otherwise, return to step S402, where the maximum number of iterations is a custom parameter.

[0066] The present invention provides a smart sanitation operation planning method, comprising the following steps:

[0067] Step S501, collecting regional characteristic data of M sub-areas of the area to be cleaned and vehicle characteristic data of N sanitation vehicles;

[0068] Step S502, preprocessing the regional feature data of the M sub-regions and the vehicle feature data of the N sanitation vehicles to obtain a third feature sequence and a fourth feature sequence;

[0069] Step S503, constructing a job time prediction model, inputting the third feature sequence and the fourth feature sequence, and outputting an estimated completion time;

[0070] Step S504, generating a planning scheme according to the operation time prediction model combined with the crested porcupine optimization algorithm that introduces the Cauchy mutation strategy.

[0071] The beneficial effect of the present invention is that the present invention realizes the collaborative operation of multiple sanitation vehicles by constructing an operation time prediction model combined with the crown porcupine optimization algorithm that introduces the Cauchy mutation strategy, improves the efficiency of sanitation operations, and realizes efficient resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a module schematic diagram of a smart sanitation operation planning system of the present invention;

[0073] Figure 2 is a flow chart of preprocessing the first feature sequence and the second feature sequence of the present invention;

[0074] Figure 3 is a schematic diagram of the operation time prediction model of the present invention;

[0075] Figure 4 It is a flow chart of the crown porcupine optimization algorithm of the present invention that introduces the Cauchy mutation strategy;

[0076] Figure 5 It is a flow chart of a smart sanitation operation planning method of the present invention.

[0077] In the figure: data acquisition module 101, data preprocessing module 102, operation time prediction model construction module 103, planning scheme generation module 104, first hidden layer 301, second hidden layer 302, separation layer 303, combination layer 304, classifier 305. DETAILED DESCRIPTION

[0078] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0079] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in one or more embodiments of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0080] like Figure 1 to Figure 5 As shown, a smart sanitation operation planning system includes:

[0081] A data collection module 101 is used to collect regional characteristic data of M sub-areas of the area to be cleaned and vehicle characteristic data of N sanitation vehicles;

[0082] Regional characteristic data include: traffic flow index, number of traffic lights, number of garbage collection points, number of vehicles, total length of roads and area of ​​the region;

[0083] Vehicle characteristic data include: average travel speed, loading efficiency, unloading efficiency and maximum loading capacity;

[0084] The regional feature data of the M sub-regions are represented by the first feature sequence, and the regional feature data of the m-th sub-region is represented by the m-th sequence unit of the first feature sequence, where 1≤m≤M;

[0085] The vehicle characteristic data of N sanitation vehicles are represented by the second characteristic sequence, and the vehicle characteristic data of the nth sanitation vehicle is represented by the nth sequence unit of the second characteristic sequence, where 1≤n≤N;

[0086] A data preprocessing module 102, which is used to preprocess the first feature sequence and the second feature sequence to obtain a third feature sequence and a fourth feature sequence;

[0087] An operation time prediction model building module 103, which is used to build an operation time prediction model according to the third feature sequence and the fourth feature sequence;

[0088] The input of the job time prediction model is the third feature sequence and the fourth feature sequence, and the output is the estimated completion time;

[0089] A planning scheme generating module 104 is used to generate a planning scheme according to the operation time prediction model combined with the crested porcupine optimization algorithm that introduces the Cauchy mutation strategy;

[0090] According to the planning scheme, M sub-areas are allocated to N sanitation vehicles.

[0091] In one embodiment of the present invention, M, N, m and n are all user-defined parameters, where M>N.

[0092] In one embodiment of the present invention, the traffic flow index is a quantitative indicator used to measure the traffic congestion of the road where the collection point is located. The formula of the traffic flow index of the mth sub-area is: ,in represents the traffic flow index of the mth sub-area, represents the number of sanitation vehicles in the mth sub-area, represents the area of ​​the mth sub-region;

[0093] The loading efficiency of a sanitation truck refers to the amount of garbage that the sanitation truck can load in one minute, where the time required for loading does not include the time the sanitation truck is driving on the road. The formula for the loading efficiency of the nth sanitation truck is: ,in represents the loading efficiency of the nth sanitation vehicle, represents the maximum loading capacity of the nth sanitation vehicle, represents the average time it takes for the nth sanitation truck to be fully loaded with garbage;

[0094] The unloading efficiency of a sanitation truck refers to the amount of garbage that can be unloaded by the sanitation truck within one minute. The time required for unloading does not include the time the sanitation truck is driving on the road. The formula for the unloading efficiency of the nth sanitation truck is: ,in represents the unloading efficiency of the nth sanitation vehicle, It represents the average time for the nth sanitation truck to unload a full load of garbage.

[0095] In one embodiment of the present invention, Figure 2 As shown, the step of preprocessing the first feature sequence and the second feature sequence includes:

[0096] Step S201, for missing values ​​in the first characteristic sequence and the second characteristic sequence, filling processing is performed by taking the average value of the value of the previous sequence unit and the value of the next sequence unit;

[0097] Step S202, removing duplicate values ​​in the first feature sequence and the second feature sequence;

[0098] Step S203: normalize the first feature sequence and the second feature sequence using a z-score normalization method.

[0099] In one embodiment of the present invention, Figure 3 As shown, the operation time prediction model includes: a first hidden layer, a second hidden layer, a separation layer, a combination layer and a classifier;

[0100] The first hidden layer inputs the third feature sequence and outputs the first updated feature;

[0101] The second hidden layer includes N hidden units, the nth hidden unit inputs the nth sequence unit of the fourth feature sequence, and outputs the second hidden feature;

[0102] The separation layer is used to separate the second hidden feature output by the Nth hidden unit as the second update feature;

[0103] The combination layer is used to expand the first update feature into a vector representation and concatenate it with the second update feature to obtain a combined vector;

[0104] The input of the classifier is a combination vector, and the classification space of the classifier represents the estimated completion time.

[0105] In one embodiment of the present invention, the formula of the first hidden layer is:

[0106] ;

[0107] ;

[0108] Where A represents the first updated feature of the first hidden layer output, e represents the attention score, and denote the first weight parameter and the second weight parameter respectively, represents the first weight vector, Q, K and V represent the first intermediate feature, the second intermediate feature and the third intermediate feature of the first hidden layer, The number of dimensions of each sequence unit representing the third characteristic sequence is a constant value of 6, and T represents a transposition operation.

[0109] In one embodiment of the present invention, the second hidden layer is constructed based on a gated recurrent unit. The second hidden layer may also be constructed based on a long short-term memory network unit or a recurrent neural unit.

[0110] In one embodiment of the present invention, MATLAB / Simulink simulation software is selected to realize the simulation of sanitation vehicle operation planning and garbage collection process. Define an area to be cleaned, divide the area to be cleaned into several sub-areas, define regional characteristic data for each sub-area, such as traffic flow index, number of traffic lights, number of garbage collection points, number of vehicles, total road length and area, define vehicle characteristic data of sanitation vehicles, such as average driving speed, loading efficiency, unloading efficiency and maximum loading capacity, and set system rules, for example: in a sub-area, the sanitation vehicle selects the next garbage collection point according to the principle of the shortest distance; the sanitation vehicle must move within a sub-area each time. Use MATLAB / Simulink to run simulation experiments of multiple scenarios, obtain simulation results, and calculate the time for each vehicle to operate in each sub-area based on the simulation results as the training label of the operation time prediction model.

[0111] In one embodiment of the present invention, Figure 4 As shown, the steps of the crested porcupine optimization algorithm using the Cauchy mutation strategy used in the planning scheme generation module include:

[0112] Step S401, generating an initialization population with P number of individuals meeting the constraint conditions, where P is a custom parameter;

[0113] The constraints include: each sanitation vehicle is assigned to at least one sub-area; each sub-area can only be assigned to one sanitation vehicle; the sub-areas assigned to each sanitation vehicle must be adjacent to ensure that multiple sanitation vehicles do not appear on the same road and cause traffic jams.

[0114] The individuals of the initialized population are represented by matrix coding, which is expressed as: ,in They represent the allocation of the first sub-area to the Mth sub-area to the first sanitation vehicle, They represent the allocation of the first sub-area to the first sanitation vehicle to the Nth sanitation vehicle, respectively. Indicates the allocation of the Mth sub-area to the Nth sanitation vehicle, and the element value of the matrix code is represented by 0 or 1;

[0115] Step S402, constructing an objective function by combining the operation time prediction model to calculate and obtain the fitness value of each individual in the initialization population;

[0116] The objective function is calculated as follows: , where Fitness represents the fitness value of the individual, It means that the regional characteristic data of the mth sub-region and the vehicle characteristic data of the nth sanitation vehicle are input into the estimated completion time output by the operation time prediction model. Represents the weight parameter of the objective function, which is a custom parameter;

[0117] Step S403, generating a random number ranging from 0 to 1, and when the random number is less than a first threshold, performing global exploration, otherwise performing local development, wherein the first threshold is a custom parameter, and preferably, the first threshold is set to 0.8;

[0118] Step S4041: during global exploration, a random number ranging from 0 to 1 is generated. When the random number is less than a second threshold, the first defense strategy is adopted; otherwise, the second defense strategy is adopted. The first threshold is a custom parameter. Preferably, the first threshold is set to 0.8.

[0119] Step S4042: during local development, a random number ranging from 0 to 1 is generated. When the random number is less than the second threshold, the third defense strategy is adopted, otherwise the fourth defense strategy is adopted.

[0120] The formula for the first defense strategy is:

[0121] ;

[0122] ;

[0123] The formula for the second defense strategy is:

[0124] ;

[0125] The formula for the third defense strategy is:

[0126] ;

[0127] ;

[0128] ;

[0129] ;

[0130] The formula for the fourth defense strategy is:

[0131] ;

[0132] in represents the position of the i-th individual when the number of iterations is k+1, represents the position of the i-th individual when the number of iterations is k, represents random numbers that follow a normal distribution, , , , and rand is a random number between 0 and 1, represents the global optimal solution when the number of iterations is k, represents the position of the predator when the number of iterations is k, r is a random integer between 1 and P, represents a binary vector consisting entirely of 0s, then , the predator no longer moves towards the crested porcupine, when When contains 1, the second defense strategy is based on two random individuals and To determine whether the predator is approaching or moving away from the crested porcupine, r1, r2, and r3 are random integers between 1 and P. represents the first vector parameter used to control the search direction, represents the defense factor when the number of iterations is k, It represents the odor diffusion factor when the number of iterations is k, where k is the current number of iterations. is the maximum number of iterations, represents the fitness of the i-th individual when the number of iterations is k, For the second parameter, avoid division by 0. is the speed convergence factor, exp represents the exponential function;

[0133] Step S405, mutating the individuals in the population based on the Cauchy mutation strategy;

[0134] The probability density function of the Cauchy distribution is: ,in represents the location parameter, γ represents the scale parameter, and x represents the custom variable;

[0135] By introducing the Cauchy mutation strategy, the search capability and convergence speed of the crested porcupine algorithm can be effectively improved;

[0136] Step S4051, determine the location parameter and scale parameter of the Cauchy distribution, preferably, the location parameter Set to 0, the scale parameter γ is adjusted according to the current number of iterations, ,in represents the scale parameter when the number of population iterations is k+1, represents the scale parameter when the number of population iterations is k, β represents a decreasing factor between 0 and 1, preferably, the decreasing factor is set to 0.99 and the scale parameter is set to 0.7;

[0137] Step S4052, determining the variation ratio, preferably, the variation ratio is set to 10%;

[0138] Step S4053, generating a mutant individual, for the selected individual, randomly extracting a mutation value from the Cauchy distribution, and applying it to each element value of the individual, the specific application method is: when the mutation value is greater than the third threshold, flipping the element value mapped by the mutation value, that is, 0 becomes 1, 1 becomes 0, preferably, the third threshold is set to 0.5;

[0139] Step S406, after the current iteration is completed, the number of populations is reduced according to the cyclic population reduction technique, and the scale parameter is updated;

[0140] The formula for the cyclic population reduction technique is:

[0141] ;

[0142] in represents the number of individuals in the updated population, Represents the minimum value of the population, preferably, Set to 10, represents the population size when the number of iterations is k, Represents a loop variable, preferably, Set to 2, Represents the remainder operation;

[0143] Step S407, when the current number of iterations is greater than or equal to the maximum number of iterations, the individual with the largest fitness in the initialized population is used as the planning scheme, otherwise return to step S402, where the maximum number of iterations is a custom parameter, preferably, the maximum number of iterations is set to 200.

[0144] In one embodiment of the present invention, Figure 5 As shown, a smart sanitation operation planning method is provided, comprising the following steps:

[0145] Step S501, collecting regional characteristic data of M sub-areas of the area to be cleaned and vehicle characteristic data of N sanitation vehicles;

[0146] Step S502, preprocessing the regional feature data of the M sub-regions and the vehicle feature data of the N sanitation vehicles to obtain a third feature sequence and a fourth feature sequence;

[0147] Step S503, constructing a job time prediction model, inputting the third feature sequence and the fourth feature sequence, and outputting an estimated completion time;

[0148] Step S504, generating a planning scheme according to the operation time prediction model combined with the crested porcupine optimization algorithm that introduces the Cauchy mutation strategy.

[0149] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are within the protection of the present embodiment.

Claims

1. A smart sanitation operation planning system, characterized in that: include: A data acquisition module, which is used to collect regional characteristic data of M sub-areas of the area to be cleaned and vehicle characteristic data of N sanitation vehicles; Regional characteristic data include: traffic flow index, number of traffic lights, number of garbage collection points, number of vehicles, total length of roads and area of ​​the region; Vehicle characteristic data include: average travel speed, loading efficiency, unloading efficiency and maximum loading capacity; The regional feature data of the M sub-regions are represented by the first feature sequence, and the regional feature data of the m-th sub-region is represented by the m-th sequence unit of the first feature sequence, where 1≤m≤M; The vehicle characteristic data of N sanitation vehicles are represented by the second characteristic sequence, and the vehicle characteristic data of the nth sanitation vehicle is represented by the nth sequence unit of the second characteristic sequence, where 1≤n≤N; A data preprocessing module, which is used to preprocess the first feature sequence and the second feature sequence to obtain a third feature sequence and a fourth feature sequence; An operation time prediction model building module, which is used to build an operation time prediction model according to the third feature sequence and the fourth feature sequence; The input of the job time prediction model is the third feature sequence and the fourth feature sequence, and the output is the estimated completion time; A planning scheme generation module is used to generate a planning scheme based on the operation time prediction model combined with the crown porcupine optimization algorithm that introduces the Cauchy mutation strategy; According to the planning scheme, M sub-areas are allocated to N sanitation vehicles.

2. According to claim 1, a smart sanitation operation planning system is characterized in that: M, N, m and n are all custom parameters, where M>N.

3. According to claim 1, a smart sanitation operation planning system is characterized in that: The formula for the traffic flow index of the mth sub-area is: ,in represents the traffic flow index of the mth sub-area, represents the number of sanitation vehicles in the mth sub-area, represents the area of ​​the mth sub-region; The formula for the loading efficiency of the nth sanitation vehicle is: ,in represents the loading efficiency of the nth sanitation vehicle, represents the maximum loading capacity of the nth sanitation vehicle, represents the average time it takes for the nth sanitation truck to be fully loaded with garbage; The formula for the unloading efficiency of the nth sanitation vehicle is: ,in represents the unloading efficiency of the nth sanitation vehicle, It represents the average time for the nth sanitation truck to unload a full load of garbage.

4. According to claim 1, a smart sanitation operation planning system is characterized in that: The step of preprocessing the first feature sequence and the second feature sequence comprises: Step S201, for missing values ​​in the first characteristic sequence and the second characteristic sequence, filling processing is performed by taking the average value of the value of the previous sequence unit and the value of the next sequence unit; Step S202, removing duplicate values ​​in the first feature sequence and the second feature sequence; Step S203: normalize the first feature sequence and the second feature sequence using a z-score normalization method.

5. The intelligent sanitation operation planning system according to claim 1 is characterized in that: The operation time prediction model includes: a first hidden layer, a second hidden layer, a separation layer, a combination layer and a classifier; The first hidden layer inputs the third feature sequence and outputs the first updated feature; The second hidden layer includes N hidden units, the nth hidden unit inputs the nth sequence unit of the fourth feature sequence, and outputs the second hidden feature; The separation layer is used to separate the second hidden feature output by the Nth hidden unit as the second update feature; The combination layer is used to expand the first update feature into a vector representation and concatenate it with the second update feature to obtain a combined vector; The input of the classifier is a combination vector, and the classification space of the classifier represents the estimated completion time.

6. The intelligent sanitation operation planning system according to claim 5 is characterized in that: The formula for the first hidden layer is: ； ； Where A represents the first updated feature of the first hidden layer output, e represents the attention score, and denote the first weight parameter and the second weight parameter respectively, represents the first weight vector, Q, K and V represent the first intermediate feature, the second intermediate feature and the third intermediate feature of the first hidden layer, The number of dimensions of each sequence unit representing the third characteristic sequence is a constant value of 6, and T represents a transposition operation.

7. The intelligent sanitation operation planning system according to claim 5 is characterized in that: The second hidden layer is built based on gated recurrent units.

8. The intelligent sanitation operation planning system according to claim 1 is characterized in that: A simulation platform is used to simulate the area to be cleaned and the sanitation vehicles, and training data for different scenarios and label data of corresponding instances are generated based on the simulation platform, where the label data is represented by the time when the sanitation vehicle handles the garbage collection point in the sub-area.

9. The intelligent sanitation operation planning system according to claim 1 is characterized in that: The steps of the crested porcupine optimization algorithm using the Cauchy mutation strategy used in the planning scheme generation module include: Step S401, generating an initialization population with P number of individuals meeting the constraint conditions, where P is a custom parameter; The constraints include: each sanitation vehicle is assigned to at least one sub-area; each sub-area can only be assigned to one sanitation vehicle; the sub-areas assigned to each sanitation vehicle must be adjacent; The individuals of the initialized population are represented by matrix coding, which is expressed as: ,in They represent the allocation of the first sub-area to the Mth sub-area to the first sanitation vehicle, They represent the allocation of the first sub-area to the first sanitation vehicle to the Nth sanitation vehicle, respectively. Indicates the allocation of the Mth sub-area to the Nth sanitation vehicle, and the element value of the matrix code is represented by 0 or 1; Step S402, constructing an objective function by combining the operation time prediction model to calculate and obtain the fitness value of each individual in the initialization population; The objective function is calculated as follows: , where Fitness represents the fitness value of the individual, It means that the regional characteristic data of the mth sub-region and the vehicle characteristic data of the nth sanitation vehicle are input into the estimated completion time output by the operation time prediction model. Represents the weight parameter of the objective function, which is a custom parameter; Step S403, generating a random number ranging from 0 to 1, and when the random number is less than a first threshold, performing global exploration, otherwise performing local development, wherein the first threshold is a custom parameter; Step S4041, during global exploration, a random number ranging from 0 to 1 is generated. When the random number is less than a second threshold, the first defense strategy is adopted, otherwise the second defense strategy is adopted, wherein the first threshold is a custom parameter; Step S4042: during local development, a random number ranging from 0 to 1 is generated. When the random number is less than the second threshold, the third defense strategy is adopted, otherwise the fourth defense strategy is adopted. Step S405, mutating the individuals in the population based on the Cauchy mutation strategy; The probability density function of the Cauchy distribution is: ,in represents the location parameter, γ represents the scale parameter, and x represents the custom variable; Step S4051, determine the location parameter and scale parameter of the Cauchy distribution, and adjust the scale parameter γ according to the current number of iterations. ,in represents the scale parameter when the number of population iterations is k+1, represents the scale parameter when the number of population iterations is k, and β represents a decreasing factor between 0 and 1; Step S4052, determining the variation ratio of individuals in the population; Step S4053, generating a mutant individual, for the selected individual, randomly extracting a mutation value from the Cauchy distribution, and applying it to each element value of the individual, the specific application method is: when the mutation value is greater than the third threshold, flipping the element value mapped by the mutation value, that is, 0 becomes 1, and 1 becomes 0; Step S406, after the current iteration is completed, the number of populations is reduced according to the cyclic population reduction technique, and the scale parameter is updated; Step S407, when the current number of iterations is greater than or equal to the maximum number of iterations, the individual with the largest fitness in the initialized population is used as the planning scheme, otherwise, return to step S402, where the maximum number of iterations is a custom parameter.

10. A smart sanitation operation planning method, characterized in that: A smart sanitation operation planning system as described in any one of claims 1 to 9 is used, comprising the following steps: Step S501, collecting regional characteristic data of M sub-areas of the area to be cleaned and vehicle characteristic data of N sanitation vehicles; Step S502, preprocessing the regional feature data of the M sub-regions and the vehicle feature data of the N sanitation vehicles to obtain a third feature sequence and a fourth feature sequence; Step S503, constructing a job time prediction model, inputting the third feature sequence and the fourth feature sequence, and outputting an estimated completion time; Step S504, generating a planning scheme according to the operation time prediction model combined with the crested porcupine optimization algorithm that introduces the Cauchy mutation strategy.

Images