Urban environmental sanitation operation optimization method based on big data

Abstract

The invention discloses an urban environmental sanitation operation optimization method based on big data, and relates to the technical field of intelligent environmental sanitation optimization, and the method comprises the steps: collecting urban environmental sanitation data, building a pollution region model, abstracting a detection process of mobile detection equipment into a Poisson random process, and carrying out the optimization of the Poisson random process; the method comprises the following steps: deducing a probability distribution function detected in a polluted area, then calculating the lowest detection frequency of each area, calculating the optimal detection frequency of each area by adopting a simulated annealing algorithm, and carrying out global planning and scheduling on a sanitation operation path by using an improved NSGA-III multi-objective optimization algorithm. Through combination of a Poisson random process model and a simulated annealing algorithm, scientific quantification and optimal distribution of detection frequency are realized, the purpose of accurate detection is achieved, the utilization rate of detection equipment is effectively improved, the regional detection completion degree is improved by using the improved NSGA-III, and meanwhile, resource consumption of invalid operation is reduced. And the environmental sanitation task execution is more efficient.

Description

Technical Field

[0001] The present invention relates to the field of intelligent sanitation optimization technology, and in particular to a method for optimizing urban sanitation operations based on big data. Background Art

[0002] With the acceleration of urbanization, the scheduling and optimization of resources required for urban sanitation has become a complex, systemic problem. Traditional sanitation operations typically rely on manual experience and fixed execution plans, making it difficult to dynamically adjust tasks based on regional pollution conditions or rationally allocate resources to cover the city's ever-changing pollution zones. In recent years, the widespread application of big data technology has provided more precise means for urban management and control. By collecting and analyzing real-time environmental data, it is possible to understand the dynamic mechanisms of urban pollution at both macro and micro levels, and optimize sanitation tasks and equipment deployment. While existing technologies have made some progress in pollution monitoring, task statistics, and route planning, they still suffer from numerous shortcomings, making it difficult to strike a balance between efficient operations and resource conservation in the urban sanitation sector. For example, traditional route planning algorithms often ignore the dynamic demands of task priorities and regional inspection frequency, relying solely on static geographic information resources. This results in relatively fixed route designs that are unable to adapt to rapidly changing pollution conditions. Furthermore, equipment fault monitoring remains largely a manual process, with limited research on fault prediction modeling and automated decision support, impacting the reliability and scheduling efficiency of sanitation equipment.

[0003] Existing technologies focus on the following aspects: First, in the processing and analysis of pollution monitoring data, traditional methods are often based on simple pollution index thresholds and lack modeling methods based on stochastic process theory. This makes it impossible to scientifically evaluate the probability of regional pollution detection, and thus presents difficulties in calculating the minimum detection frequency and optimizing the optimal detection frequency. Second, in sanitation route planning algorithms, some studies have introduced multi-objective optimization methods, but these methods fail to fully consider the dynamic scheduling requirements and the importance of regional tasks in route optimization. In particular, in vehicle scheduling, there is no dynamic adjustment mechanism based on regional pollution models, and the potential of modern optimization algorithms is not fully utilized. Third, for the detection and maintenance of sanitation equipment failures, the simple fault alarm model currently used cannot effectively predict the probability of failure and adjust task plans in a timely manner, often resulting in the participation of high-risk equipment in tasks, further deteriorating operational conditions. Furthermore, the assessment of equipment deployment scale lacks a quantitative method to balance cost and efficiency, making it impossible to ensure the optimal deployment of equipment resources. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an urban sanitation operation optimization method based on big data to solve the problems of inaccurate pollution judgment, inaccurate detection frequency and low path scheduling efficiency in existing urban sanitation operations.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for optimizing urban sanitation operations based on big data, comprising collecting and preprocessing urban sanitation data, determining pollution status labels for each area based on the preprocessed data, including pollution status, pollution type, and risk level, constructing a pollution area model based on the determination results, generating multiple types of sanitation operation tasks based on the pollution status labels for each area in the pollution area model, and adjusting the pollution status labels of the pollution area model based on the completion status of the operation;

[0008] The pollution detection process of mobile detection equipment is abstracted as a Poisson random process. The probability distribution function of the detected polluted area is derived, and the minimum detection frequency of each area is calculated. Based on the regional minimum detection frequency, the simulated annealing algorithm is used to calculate the optimal detection frequency of each area. Based on the regional optimal detection frequency, the improved NSGA-III multi-objective optimization algorithm is used to perform global planning and scheduling of sanitation operation routes.

[0009] During sanitation operation, the LSTM prediction model is used to predict the failure probability of sanitation equipment, screen out high-risk equipment from participating in route scheduling, and automatically generate maintenance orders. At the same time, the cost-efficiency function method is used to calculate the deployment scale critical point, and sanitation equipment is deployed based on the calculation results.

[0010] As a preferred solution of the urban sanitation operation optimization method based on big data described in the present invention, wherein: the improved NSGA-III multi-objective optimization algorithm is used based on the regional optimal detection frequency to perform global planning and scheduling of the sanitation operation path. Generate N scheduling plans, each of which is represented by S, and the task allocation of each area must meet its optimal detection frequency

[0011] Calculate the optimization objective function value for each scheduling scheme, including the total path length minimization function f1(S), the detection coverage maximization function f2(S), the task balance maximization function f3(S), and the operation cost minimization function f4(S). The multi-objective optimization calculation result is recorded as the objective function value F(S);

[0012] According to the objective function value F(S), all the solutions in the solution set are Pareto sorted. If solution S i It is not worse than solution S in all optimization objectives. j, and Scheme S i Strictly better than solution S in at least one optimization objective j , then Scheme S i Pareto dominance solution S j ;

[0013] Calculate the sparsity d of the set of non-dominated solutions crowd (S i ), set the sparsity threshold If d crowd (S i ) is greater than or equal to Then plan S i Put it into the set of preferred solutions, otherwise it will not be adopted;

[0014] Based on the selected optimal solution set, crossover and mutation operations are performed to randomly select two solutions from the optimal solution set, and some task scheduling of the two solutions is exchanged to generate a new sub-solution S new , and randomly select an indefinite number of preferred solutions to meet the regional detection frequency Under the condition of , the task allocation of the scheme is randomly adjusted to generate new sub-schemes, and the generated sub-schemes are integrated with the optimal scheme to obtain a new candidate scheme set W new ;

[0015] Set the maximum number of iterations For set W new The schemes in the set are Pareto sorted, and the sparsity of the non-dominated schemes obtained after sorting is calculated. The non-dominated schemes with a sparsity greater than the set threshold are set as the new set of preferred schemes. Before the number of iterations reaches the threshold, new sub-schemes are repeatedly generated, and a new set of preferred schemes is obtained after Pareto sorting and sparsity calculation.

[0016] When the number of iterations reaches the upper limit Then, output the non-dominated and sparsest solution S among the preferred solutions. best as the optimal scheduling solution.

[0017] As a preferred embodiment of the method for optimizing urban sanitation operations based on big data described in the present invention, generating multiple types of sanitation operation tasks based on the pollution status labels of each area in the pollution area model and adjusting the pollution status labels of the pollution area model based on the completion status of the operation refers to reading the pollution status, pollution type, and risk level of each area in the pollution area model, and determining whether to generate a sanitation operation task based on the read content:

[0018] If the pollution status is "polluted", the sanitation task will be generated, otherwise it will not be generated;

[0019] If the pollution type is particulate matter, a watering and dust reduction task is generated; if it is odorous, a spraying and deodorization task is generated; if it is piled, a garbage removal task is generated;

[0020] Equivalently equate risk levels to task priorities, and sort the generated sanitation tasks by task priority and incorporate them into job scheduling;

[0021] The task completion status is uploaded through the vehicle terminal. If the task is completed, the pollution status of the corresponding area will be set to "no pollution". If the task is not completed, the pollution status will be maintained and sanitation operation tasks will continue to be generated.

[0022] As a preferred solution of the urban sanitation operation optimization method based on big data described in the present invention, the pollution detection process of the mobile detection equipment is abstracted into a Poisson random process, and the probability distribution function of the polluted area is derived to calculate the minimum detection frequency of each area, which means calculating the number of times the i-th area needs to be detected in unit time λ i ;

[0023] The detection behavior of mobile pollution detection equipment in each area is modeled as a Poisson distribution P;

[0024] Calculate the probability that the i-th region is detected at least once within the time period ρ based on the Poisson distribution

[0025] Set the probability threshold θ that the i-th region is detected at least once within the time interval ρ, and calculate the minimum required detection frequency

[0026] As a preferred solution of the urban sanitation operation optimization method based on big data described in the present invention, wherein: the optimal detection frequency index of each area is calculated by using the simulated annealing algorithm based on the regional minimum detection frequency. Initialize to the current detection frequency

[0027] Based on energy cost function Response efficiency function Calculate the current detection frequency The optimization objective function value under

[0028] Generate candidate detection frequencies in the neighborhood of the current detection frequency and Greater than

[0029] Calculate the optimization objective function value of the candidate frequency Compare the optimization objective function value of the candidate detection frequency with the current detection frequency. If Greater than or equal to Then keep the current detection frequency. If Less than Then the candidate detection frequency is set as the current detection frequency;

[0030] Calculate the difference between the optimization objective function value of the current detection frequency and the candidate detection frequency

[0031] Set the objective function change threshold like Greater than or equal to Then continue to generate candidate detection frequencies. If Less than It indicates that it has converged to the optimal frequency and output is the optimal detection frequency.

[0032] As a preferred solution of the big data-based urban sanitation operation optimization method described in the present invention, the LSTM prediction model is used to predict the failure probability of sanitation equipment during the sanitation operation process, screen out high-risk equipment from participating in route scheduling, and automatically generate maintenance orders to record equipment information once an hour and output the state sequence X of the equipment at the last n moments;

[0033] Based on the equipment's state sequence, the LSTM model is used to predict the probability ξ of equipment failure within the next τ time. A failure probability threshold is set. If the predicted failure probability is greater than or equal to the failure probability threshold, the equipment is identified as high-risk equipment and does not participate in the path planning of sanitation work. A maintenance order is directly generated for equipment maintenance.

[0034] As a preferred solution of the urban sanitation operation optimization method based on big data described in the present invention, wherein: the cost-efficiency function method is used to calculate the critical point of deployment scale, and the sanitation equipment deployment index is constructed based on the calculation results. Cost function of a device Building efficiency functions Calculating unit efficiency costs right Derivative, solve the minimum point where the derivative is zero as the optimal deployment Use the calculated As the optimal number of devices to deploy.

[0035] As a preferred solution of the big data-based urban sanitation operation optimization method described in the present invention, the steps of collecting and preprocessing urban sanitation data, determining pollution status labels of various areas based on the preprocessed data, and constructing pollution area models based on the determination results refer to deploying sensing devices at fixed monitoring stations and mobile detection equipment to collect environmental data in real time, including geographic location, particle size, environmental images, abnormal gas concentrations, and equipment numbers, and uploading the data to a database after preprocessing.

[0036] The pollution status label includes the pollution status, pollution type and risk level. The threshold of particle size in the air, the threshold of abnormal gas concentration and the threshold of solid waste area are set. The data in the database are compared with the set thresholds. If the particle size is large, the abnormal gas concentration or the solid waste area identified by the image is greater than or equal to the set threshold, the pollution status of the area is marked as "polluted" and the three pollution types are recorded as particulate matter type, odor type and pile type respectively. Otherwise, the pollution status of the area is marked as "non-polluted";

[0037] Normalize and calculate the pollution index value A, and calculate the pollution risk level value RiskScore of the i-th area based on the pollution index value and the duration of pollution i ;

[0038] Set a low risk threshold RiskScore low and high risk threshold RiskScore high :

[0039] If RiskScore i Less than RiskScore low , then the risk level of the i-th area is set to level one;

[0040] If RiskScore i Greater than or equal to RiskScore low and less than or equal to RiskScore high , then the risk level of the i-th area is set to level 2;

[0041] If RiskScore i Greater than RiskScore high , then the risk level of the i-th area is set to level three;

[0042] The city is divided into a grid model of size m, where nodes represent regions and edges represent adjacency relationships. The pollution status label of each region is embedded in the grid model to form a pollution area model.

[0043] In a second aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the urban sanitation operation optimization method based on big data as described in the first aspect of the present invention.

[0044] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of a method for optimizing urban sanitation operations based on big data as described in the first aspect of the present invention.

[0045] The beneficial effects of the present invention are as follows: through the combination of the Poisson random process model and the simulated annealing algorithm, the scientific quantification and optimal allocation of the detection frequency are realized, which not only achieves the purpose of accurate detection, but also effectively improves the utilization rate of the detection equipment, and reduces the cost of detection resource allocation; the use of the improved NSGA-III can realize the dynamic and real-time performance of path planning, improve the completion rate of regional detection, and reduce the resource consumption of ineffective operation, making the execution of sanitation tasks more efficient; the use of the LSTM model to predict the probability of equipment failure and optimize equipment scheduling realizes the intelligentization of fault prediction and maintenance automation decision-making, significantly improves the utilization efficiency of sanitation equipment, and further reduces the cost of equipment deployment through scale optimization, thereby improving the reliability of task execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0047] Figure 1 This is a flow chart of a method for optimizing urban sanitation operations based on big data in Example 1.

[0048] Figure 2 This is a core flow chart of a method for optimizing urban sanitation operations based on big data in Example 1. DETAILED DESCRIPTION

[0049] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0050] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0051] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0052] Example 1, with reference to Figure 1 and Figure 2, which is the first embodiment of the present invention, provides a method for optimizing urban sanitation operations based on big data, comprising the following steps:

[0053] S1: Collect urban sanitation data and preprocess it. Determine the pollution status labels of each area based on the preprocessed data, including pollution status, pollution type, and risk level. Build a pollution area model based on the determination results. Generate multiple types of sanitation operation tasks based on the pollution status labels of each area in the pollution area model. Adjust the pollution status labels of the pollution area model based on the completion status of the operation.

[0054] Specifically, it collects and pre-processes urban sanitation data, determines the pollution status label of each area based on the pre-processed data, and builds a pollution area model based on the determination results. This means deploying a variety of sensing devices on fixed monitoring stations and mobile detection equipment, including particulate matter sensors, image acquisition devices, and gas detection devices, to collect environmental data in real time, including geographic location, particle size, environmental images, abnormal gas concentrations, and equipment numbers. The data is pre-processed and uploaded to the database;

[0055] The pollution status label includes the pollution status, pollution type and risk level. The threshold of particle size in the air, the threshold of abnormal gas concentration and the threshold of solid waste area are set. The data in the database are compared with the set thresholds. If the particle size is large, the abnormal gas concentration or the solid waste area identified by the image is greater than or equal to the set threshold, the pollution status of the area is marked as "polluted" and the three pollution types are recorded as particulate matter type, odor type and pile type respectively. Otherwise, the pollution status of the area is marked as "non-polluted";

[0056] Normalized calculation of pollution index value A:

[0057]

[0058] Among them, a grain 、a air 、a solid is the actual value of pollution in terms of particle size, gas concentration and solid waste area in the air, The maximum tolerance values ​​set for particle size, gas concentration, and solid waste area in the air, ψ grain , ψ air , ψ solid Weights assigned to airborne particle size, gas concentrations, and solid waste areas;

[0059] Calculate the pollution risk level value RiskScorei of the i-th area based on the pollution index value and pollution duration:

[0060] RiskScore i =ω1·A i+ω2·D i ,

[0061] Among them, A i is the normalized pollution index value of the ith region, D i is the duration of pollution in the i-th area, ω1 and ω2 are weight values ​​that can be optimized according to training;

[0062] Set a low risk threshold RiskScore low and high risk threshold RiskScore high :

[0063] If RiskScore i Less than RiskScore low , then the risk level of the i-th area is set to level one;

[0064] If RiskScore i Greater than or equal to RiskScore low and less than or equal to RiskScore high , then the risk level of the i-th area is set to level 2;

[0065] If RiskScore i Greater than RiskScore high , then the risk level of the i-th area is set to level three;

[0066] The city is divided into a grid model of size m, where nodes represent regions and edges represent adjacency relationships. The pollution status label of each region is embedded in the grid model to form a pollution area model.

[0067] By comparing the data collected in the database with the set threshold, automatic classification of pollution is achieved, which significantly improves the targeted execution of sanitation tasks. Normalized pollution index values ​​are calculated and regional risk levels are evaluated, which realizes dynamic risk analysis of polluted areas and effectively improves the efficiency of emergency response. The modeling of pollution area comprehensively refines the distribution of pollution in the city, provides an input structure with spatial correlation, and provides clear and complete data support for complex path planning.

[0068] Furthermore, based on the pollution status labels of each area in the pollution area model, multiple types of sanitation operation tasks are generated, and the pollution status labels of the pollution area model are adjusted based on the completion status of the operation. This refers to reading the pollution status, pollution type and risk level of each area in the pollution area model, and determining whether to generate a sanitation operation task based on the read content:

[0069] If the pollution status is "polluted", the sanitation task will be generated, otherwise it will not be generated;

[0070] If the pollution type is particulate matter, a watering and dust reduction task is generated; if it is odorous, a spraying and deodorization task is generated; if it is piled, a garbage removal task is generated;

[0071] Equivalently equate risk levels to task priorities, and sort the generated sanitation tasks by task priority and incorporate them into job scheduling;

[0072] The task completion status is uploaded through the vehicle terminal. If the task is completed, the pollution status of the corresponding area will be set to "no pollution". If the task is not completed, the pollution status will be maintained and sanitation operation tasks will continue to be generated.

[0073] By judging the pollution status of the area, the sanitation operation tasks can be accurately triggered, avoiding unnecessary waste of resources. The pollution type is judged to ensure that the task type matches the pollution situation, which improves the professionalism and effectiveness of pollution treatment. Sanitation operations are sorted and scheduled according to the risk level, and tasks in high-risk areas are given priority, which improves the efficiency and timeliness of task execution. The pollution area model is updated in real time through the completion status of the operation, which realizes closed-loop management of the pollution status and strengthens the reliability of the entire sanitation task scheduling.

[0074] S2: The pollution detection process of mobile detection equipment is abstracted as a Poisson random process. The probability distribution function of the detected polluted area is derived, and the minimum detection frequency of each area is calculated. Based on the regional minimum detection frequency, the simulated annealing algorithm is used to calculate the optimal detection frequency of each area. Based on the regional optimal detection frequency, the improved NSGA-III multi-objective optimization algorithm is used to perform global planning and scheduling of sanitation operation routes.

[0075] Specifically, the pollution detection process of the mobile detection equipment is abstracted as a Poisson random process. After the probability distribution function of the contaminated area being detected is derived, the minimum detection frequency of each area is calculated, which means the number of times the i-th area needs to be detected in unit time is calculated as λ i :

[0076]

[0077] Among them, λ i is the number of times the i-th region needs to be detected, λ0 is the set global basic detection frequency, The risk level weight is set;

[0078] The detection behavior of mobile pollution detection equipment in each area is modeled as a Poisson distribution P:

[0079]

[0080] Among them, N i(ρ) is the number of times the i-th region is detected within the time interval ρ, z is the number of detection events, and ρ is the set fixed time;

[0081] Calculate the probability that the i-th region is detected at least once within the time period ρ based on the Poisson distribution

[0082]

[0083] Set the probability threshold θ that the i-th region is detected at least once within the time interval ρ, then:

[0084]

[0085] Calculate the minimum required testing frequency

[0086]

[0087] By calculating the number of times each area needs to be inspected per unit time, scientific demand quantification is carried out in combination with regional pollution, and the regional pollution detection behavior is modeled as a Poisson process. This provides a scientific and quantifiable method for calculating key probability functions in the detection process. Based on the Poisson distribution, the probability of a region being inspected at least once per unit time is calculated, which realizes the evaluation of the detection coverage effect, calculates the minimum detection frequency, optimizes the spatiotemporal allocation efficiency of detection resources, and improves the accuracy of detection planning.

[0088] Furthermore, based on the lowest detection frequency of the region, the simulated annealing algorithm is used to calculate the optimal detection frequency of each region, which is the initialization of the current detection frequency.

[0089]

[0090] Calculate the current detection frequency The optimization objective function value under

[0091]

[0092] in, is the energy cost function of the current detection frequency, is the response efficiency function of the current detection frequency, α F and β F They are the energy consumption cost weight and response efficiency weight set based on sanitation operation demand and budget constraints;

[0093] The energy cost function Specifically expressed as:

[0094]

[0095] Among them, v i is the energy consumption cost coefficient corresponding to unit frequency, b i is fixed maintenance cost;

[0096] The response performance function Specifically expressed as:

[0097]

[0098] Among them, g i is the efficiency weight coefficient;

[0099] Generate candidate detection frequencies in the neighborhood of the current detection frequency and Greater than

[0100]

[0101] Where Δλ is the disturbance generated by uniform normal distribution;

[0102] Calculate the optimization objective function value of the candidate frequency

[0103]

[0104] Compare the optimization objective function value of the candidate detection frequency with the current detection frequency:

[0105] like Greater than or equal to Then maintain the current detection frequency;

[0106] like Less than Then the candidate detection frequency is set as the current detection frequency;

[0107] Calculate the difference between the optimization objective function value of the current detection frequency and the candidate detection frequency

[0108]

[0109] Set the objective function change threshold

[0110] like Greater than or equal to Then continue to generate candidate detection frequencies;

[0111] like Less than It indicates that it has converged to the optimal frequency and output is the optimal detection frequency.

[0112] By designing the energy cost function and the response effectiveness function, the optimization objectives of resource cost and detection efficiency are constructed, ensuring that the optimization of detection frequency is both efficient and economical. Candidate frequencies are generated in the frequency neighborhood to calculate the optimization target value, which enhances the diversity of the optimization search and greatly improves the possibility of escaping the local optimal solution. The optimization target values ​​are compared and high-quality detection frequencies are retained to ensure that the objective function tends to the best global optimal detection frequency in each round.

[0113] Furthermore, the improved NSGA-III multi-objective optimization algorithm is used to perform global planning and scheduling of sanitation operation paths based on the regional optimal detection frequency. Generate N scheduling plans, each of which is represented by S:

[0114] S={T1:[R1,R2,...],T2:[R i ,R i+1 ],...,T j :[R m-1 ,R m ]},

[0115] Among them, T j is the jth sanitation vehicle, R i is the i-th area that the sanitation vehicle is responsible for detecting, m is the number of gridded areas in the pollution area model, and the task allocation of each area must meet its optimal detection frequency

[0116] Calculate the optimization objective function value for each scheduling scheme, including total path length, detection coverage, task balance and operating cost:

[0117] The total path length minimization function f1(S):

[0118]

[0119] Among them, G j is the path of the j-th sanitation vehicle, is the total number of sanitation vehicles, p k 、p l is the task point, y(p k ,p l ) is the distance between task points;

[0120] Detection coverage maximization function f2(S):

[0121]

[0122] Among them, δ i is the indicator function, if the detection frequency of the i-th region reaches Then δi =1, otherwise δ i =0;

[0123] Task balance maximization function f3(S):

[0124]

[0125] in, is the working time of the jth sanitation vehicle, VAR is the variance of working time;

[0126] Run the cost minimization function f4(S):

[0127]

[0128] in, is the driving cost of sanitation vehicles, the resource cost of completing the mission for the vehicle;

[0129] The multi-objective optimization calculation result is recorded as the objective function value F(S):

[0130] F(S)=[f1(S),f2(S),f3(S),f4(S)],

[0131] Perform Pareto sorting on all the solutions in the solution set according to the objective function value F(S).

[0132]

[0133] in, For the plan An optimization goal, if solution S i It is not worse than solution S in all optimization objectives. j , and Scheme S i Strictly better than solution S in at least one optimization objective j , then Scheme S i Pareto dominance solution S j ;

[0134] Calculate the sparsity d of the set of non-dominated solutions crowd (S i ):

[0135]

[0136] in, is the total number of optimization objectives in the scheme, and Plan S i In the The target values ​​of the left neighbor solution and the right neighbor solution on the optimization target;

[0137] Set the sparsity threshold If d crowd (S i ) is greater than or equal to Then plan S i Put it into the set of preferred solutions, otherwise it will not be adopted;

[0138] Based on the selected optimal solution set, crossover and mutation operations are performed to randomly select two solutions S from the optimal solution set. i and S j , exchange some of the task scheduling of the two schemes to generate a new sub-scheme S new , and randomly select an indefinite number of preferred solutions to meet the regional detection frequency Under the condition of , the task allocation of the scheme is randomly adjusted to generate new sub-schemes, and the generated sub-schemes are integrated with the optimal scheme to obtain a new candidate scheme set W new ;

[0139] Set the maximum number of iterations For set W new The schemes in the set are Pareto sorted, and the sparsity of the non-dominated schemes obtained after sorting is calculated. The non-dominated schemes with a sparsity greater than the set threshold are set as the new set of preferred schemes. Before the number of iterations reaches the threshold, new sub-schemes are repeatedly generated, and a new set of preferred schemes is obtained after Pareto sorting and sparsity calculation.

[0140] When the number of iterations reaches the upper limit Then, output the non-dominated and sparsest solution S among the preferred solutions. best as the optimal scheduling solution.

[0141] By generating a scheduling plan based on the optimal regional detection frequency, it is ensured that the planning of sanitation tasks meets the detection needs of each region, optimizes the accuracy of resource allocation, and calculates the multi-objective optimization values ​​of total path length, coverage, task balance and operating cost, so that the plan reaches the multi-objective comprehensive optimality. Through Pareto sorting and sparsity calculation, non-dominated plans are screened and the distribution performance of the scheduling plan is optimized, which enhances the representativeness and diversity of the set of preferred plans. By using crossover and mutation operations, the plan optimization is freed from the local optimal limitation and the global optimization ability of the final plan is enhanced. The quality of the plan is further improved through iterative optimization. Finally, the non-dominated plan with the largest sparsity is output as the final plan, ensuring that the plan achieves a balance in multi-dimensional optimization goals and has high efficiency and low cost in executing tasks.

[0142] S3: During sanitation operation, the LSTM prediction model is used to predict the failure probability of sanitation equipment, screen out high-risk equipment from participating in route scheduling, and automatically generate maintenance orders. At the same time, the cost-efficiency function method is used to calculate the deployment scale critical point, and sanitation equipment is deployed based on the calculation results.

[0143] Specifically, during sanitation operation, the LSTM prediction model is used to predict the probability of sanitation equipment failure, screen out high-risk equipment from participating in route scheduling, and automatically generate maintenance orders to record equipment information once an hour and output the state sequence X of the equipment at the last n moments:

[0144] X=[x t-n , x t-n+1 , x t-n+2 ,...,x t ],

[0145] Among them, x t is the state of the device at time t, specifically expressed as:

[0146] x t =[Q t , O t , U t , State t ],

[0147] Among them, Q t is the temperature of the device at time t, O t is the vibration intensity of the equipment at time t, U t is the power consumption of the device at time t, State t is the abnormal status of the device at time t, State = 0 indicates that the device is in a normal state, and State = 1 indicates that the device is in an abnormal state;

[0148] Based on the state sequence of the device, the LSTM model is used to predict the probability of failure of the device within the next τ time:

[0149] ξ=LSTM(X),

[0150] Among them, ξ∈[0,1], the closer it is to 1, the higher the probability of the device failing in the next τ time;

[0151] The use of the LSTM model specifically refers to first using the LSTM_cell basic unit in LSTM to calculate the hidden state h of the device. t :

[0152] h t =LSTM_cell(x t , h t-1 ),

[0153] Among them, ht is the hidden state of the final output of the device, h t-1 is the hidden state of the device at the previous moment, and LSTM_cell is a basic computing unit in the LSTM model;

[0154] Based on the final hidden state of the device, the Sigmoid function is used to calculate the probability of the device failing within the next τ time:

[0155]

[0156] Among them, Sigmoid is the function set in the last layer of the LSTM model, κ is the weight obtained by training based on the set prediction time τ, is the bias obtained by training based on the set prediction time τ;

[0157] Set the failure probability threshold ξ θ , if ξ is greater than or equal to ξ θ , the equipment will be identified as high-risk equipment, will not participate in the path planning of sanitation work, and a maintenance order will be directly generated for equipment maintenance.

[0158] By generating the state sequence of the equipment and capturing the temporal dynamic characteristics of the equipment's operating status, a data basis is provided for fault probability prediction. The LSTM model is used to integrate the short-term and long-term data of the equipment's operation, and the hidden state of the equipment is calculated. Based on the hidden state, the Sigmoid function is used to convert the complex equipment state into an intuitive fault probability, providing a quantitative evaluation standard for equipment management. High-risk equipment is screened out from participating in sanitation tasks and repaired in advance, thus avoiding the impact of risky equipment on sanitation tasks and improving the efficiency of equipment maintenance.

[0159] Furthermore, the cost-efficiency function method is used to calculate the critical point of deployment scale, and the deployment of sanitation equipment is constructed based on the calculation results. Cost function of a device

[0160]

[0161] Among them, Γ0 is the fixed operating cost, Γ u is the cost of each device;

[0162] Building efficiency functions

[0163]

[0164] in, is the basic efficiency constant determined by the operating system structure, ι is a positive exponent representing the efficiency improvement brought by equipment deployment, and γ is an interference parameter reflecting the efficiency decline caused by excessive deployment;

[0165] Calculating unit efficiency costs

[0166]

[0167] right Derivative, solve the minimum point where the derivative is zero as the optimal deployment

[0168]

[0169] Use the calculated As the optimal number of devices to deploy.

[0170] By constructing an input cost and efficiency function, the impact of the number of equipment on cost efficiency is quantified, providing a quantitative basis for the deployment planning of sanitation equipment. Using unit efficiency cost as a measurement indicator, the unit efficiency cost is derived to find the minimum point, minimizing the cost expenditure corresponding to unit efficiency while ensuring that the deployment plan has sufficient task completion capabilities.

[0171] This embodiment also provides a computer device, which is suitable for a method for optimizing urban sanitation operations based on big data, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement a method for optimizing urban sanitation operations based on big data as proposed in the above embodiment.

[0172] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0173] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for optimizing urban sanitation operations based on big data as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0174] In summary, the present invention realizes the scientific quantification and optimal allocation of detection frequency through the combination of Poisson random process model and simulated annealing algorithm, which not only achieves the purpose of accurate detection, but also effectively improves the utilization rate of detection equipment, and reduces the cost of detection resource allocation; the use of improved NSGA-III can realize the dynamic and real-time performance of path planning, improve the completion rate of regional detection, and reduce the resource consumption of ineffective operation, making the execution of sanitation tasks more efficient; the use of LSTM model to predict the probability of equipment failure and optimize equipment scheduling realizes the intelligentization of fault prediction and maintenance automation decision-making, significantly improves the utilization efficiency of sanitation equipment, and further reduces the cost of equipment deployment through scale optimization, thereby improving the reliability of task execution.

[0175] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for optimizing urban sanitation operations based on big data, characterized by: include, Collect and preprocess urban sanitation data, determine the pollution status labels of each area based on the preprocessed data, including pollution status, pollution type, and risk level, and build a pollution area model based on the determination results. Generate multiple types of sanitation operation tasks based on the pollution status labels of each area in the pollution area model, and adjust the pollution status labels of the pollution area model based on the completion status of the operation; The pollution detection process of mobile detection equipment is abstracted as a Poisson random process. The probability distribution function of the detected polluted area is derived, and the minimum detection frequency of each area is calculated. Based on the regional minimum detection frequency, the simulated annealing algorithm is used to calculate the optimal detection frequency of each area. Based on the regional optimal detection frequency, the improved NSGA-III multi-objective optimization algorithm is used to perform global planning and scheduling of sanitation operation routes. During sanitation operation, the LSTM prediction model is used to predict the failure probability of sanitation equipment, screen out high-risk equipment from participating in route scheduling, and automatically generate maintenance orders. At the same time, the cost-efficiency function method is used to calculate the deployment scale critical point, and sanitation equipment is deployed based on the calculation results.

2. The urban sanitation operation optimization method based on big data according to claim 1, characterized in that: The improved NSGA-III multi-objective optimization algorithm is used to perform global planning and scheduling of the sanitation operation path based on the regional optimal detection frequency. Generate N scheduling plans, each of which is represented by S, and the task allocation of each area must meet its optimal detection frequency Calculate the optimization objective function value for each scheduling scheme, including the total path length minimization function f1(S), the detection coverage maximization function f2(S), the task balance maximization function f3(S), and the operation cost minimization function f4(S). The multi-objective optimization calculation result is recorded as the objective function value F(S); According to the objective function value F(S), all the solutions in the solution set are Pareto sorted. If solution S i It is not worse than solution S in all optimization objectives. j , and Scheme S i Strictly better than solution S in at least one optimization objective j , then Scheme S i Pareto dominance solution S j ; Calculate the sparsity d of the set of non-dominated solutions crowd (S i ), set the sparsity threshold If d crowd (S i ) is greater than or equal to Then plan S i Put it into the set of preferred solutions, otherwise it will not be adopted; Based on the selected optimal solution set, crossover and mutation operations are performed to randomly select two solutions from the optimal solution set, and some task scheduling of the two solutions is exchanged to generate a new sub-solution S new , and randomly select an indefinite number of preferred solutions to meet the regional detection frequency Under the condition of , the task allocation of the scheme is randomly adjusted to generate new sub-schemes, and the generated sub-schemes are integrated with the optimal scheme to obtain a new candidate scheme set W new ; Set the maximum number of iterations For set W new The schemes in the set are Pareto sorted, and the sparsity of the non-dominated schemes obtained after sorting is calculated. The non-dominated schemes with a sparsity greater than the set threshold are set as the new set of preferred schemes. Before the number of iterations reaches the threshold, new sub-schemes are repeatedly generated, and a new set of preferred schemes is obtained after Pareto sorting and sparsity calculation. When the number of iterations reaches the upper limit Then, output the non-dominated and sparsest solution S among the preferred solutions. best as the optimal scheduling solution.

3. The urban sanitation operation optimization method based on big data according to claim 2, characterized in that: Generating multiple types of sanitation operation tasks based on the pollution status labels of each area in the pollution area model and adjusting the pollution status labels of the pollution area model based on the completion status of the operation refers to reading the pollution status, pollution type and risk level of each area in the pollution area model and determining whether to generate a sanitation operation task based on the read content: If the pollution status is "polluted", the sanitation task will be generated, otherwise it will not be generated; If the pollution type is particulate matter, a watering and dust reduction task is generated; if it is odorous, a spraying and deodorization task is generated; if it is piled, a garbage removal task is generated; Equivalently equate risk levels to task priorities, and sort the generated sanitation tasks by task priority and incorporate them into job scheduling; The task completion status is uploaded through the vehicle terminal. If the task is completed, the pollution status of the corresponding area will be set to "no pollution". If the task is not completed, the pollution status will be maintained and sanitation work tasks will continue to be generated.

4. The urban sanitation operation optimization method based on big data according to claim 3, characterized in that: The pollution detection process of the mobile detection device is abstracted as a Poisson random process. After the probability distribution function of the pollution area being detected is derived, the minimum detection frequency of each area is calculated, which means the number of times the i-th area needs to be detected in unit time is calculated as λ i ; The detection behavior of mobile pollution detection equipment in each area is modeled as a Poisson distribution P; Calculate the probability that the i-th region is detected at least once within the time period ρ based on the Poisson distribution Set the probability threshold θ that the i-th region is detected at least once within the time interval ρ, and calculate the minimum required detection frequency 5. The urban sanitation operation optimization method based on big data according to claim 4, characterized in that: The optimal detection frequency index of each region is calculated by using the simulated annealing algorithm based on the lowest detection frequency of the region. Initialize to the current detection frequency Based on energy cost function Response efficiency function Calculate the current detection frequency The optimization objective function value under Generate candidate detection frequencies in the neighborhood of the current detection frequency and Greater than Calculate the optimization objective function value of the candidate frequency Compare the optimization objective function value of the candidate detection frequency with the current detection frequency. If Greater than or equal to Then keep the current detection frequency. If Less than Then the candidate detection frequency is set as the current detection frequency; Calculate the difference between the optimization objective function value of the current detection frequency and the candidate detection frequency Set the objective function change threshold. Greater than or equal to, continue to generate candidate detection frequencies, if Less than, indicating that it has converged to the optimal frequency, output is the optimal detection frequency.

6. The urban sanitation operation optimization method based on big data according to claim 5, characterized in that: During the sanitation operation, the LSTM prediction model is used to predict the probability of sanitation equipment failure, screen out high-risk equipment from participating in route scheduling, and automatically generate maintenance orders to record equipment information once an hour and output the state sequence X of the equipment at the last n moments; Based on the equipment's state sequence, the LSTM model is used to predict the probability ξ of equipment failure within the next τ time. A failure probability threshold is set. If the predicted failure probability is greater than or equal to the failure probability threshold, the equipment is identified as high-risk equipment and does not participate in the path planning of sanitation work. A maintenance order is directly generated for equipment maintenance.

7. The urban sanitation operation optimization method based on big data according to claim 6, characterized in that: The cost-efficiency function method is used to calculate the critical point of deployment scale. Deploying sanitation equipment based on the calculation result means constructing a cost function Γ(θ) for investing θ equipment, constructing an efficiency function H(θ), calculating the unit efficiency cost l(θ), taking the derivative of l(θ), and solving the minimum point where the derivative is zero as the optimal deployment θ. * , using the calculated θ * As the optimal number of devices to deploy.

8. The urban sanitation operation optimization method based on big data according to claim 7, characterized in that: The said collecting and preprocessing of urban sanitation data, determining the pollution status label of each area based on the preprocessed data, and constructing a pollution area model based on the determination results refers to deploying sensing devices on fixed monitoring stations and mobile detection equipment to collect environmental data in real time, including geographic location, particle size, environmental images, abnormal gas concentration, and equipment number, and uploading the data to the database after preprocessing; The pollution status label includes pollution status, pollution type and risk level. It sets the threshold of particle size in the air, abnormal gas concentration threshold and solid waste area threshold. The data in the database is compared with the set threshold. If the particle size is large, the abnormal gas concentration or the solid waste area identified by the image is greater than or equal to the set threshold, the pollution status of the area is marked as "polluted". The three pollution types are recorded as particulate matter type, odor type and pile type respectively. Otherwise, the pollution status of the area is marked as "unpolluted"; Normalize and calculate the pollution index value A, and calculate the pollution risk level value RiskScore of the i-th area based on the pollution index value and the duration of pollution i ; Set a low risk threshold RiskScore low and high risk threshold RiskScore high : If RiskScore i Less than RiskScore low , then the risk level of the i-th area is set to level one; If RiskScore i Greater than or equal to RiskScore low and less than or equal to RiskScore high , then the risk level of the i-th area is set to level 2; If RiskScore i Greater than RiskScore high , then the risk level of the i-th area is set to level three; The city is divided into a grid model of size m, where nodes represent regions and edges represent adjacency relationships. The pollution status label of each region is embedded in the grid model to form a pollution area model.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the urban sanitation operation optimization method based on big data according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the urban sanitation operation optimization method based on big data according to any one of claims 1 to 8 are implemented.

Images