Intelligent disinfection operation and maintenance management system and method
Through the intelligent disinfection operation and maintenance management system, using data collection and multiple algorithms to analyze and path planning, the problem of waste of resources and poor disinfection results caused by manual inspection and fixed modes in the existing technology is solved, and the problem of poor disinfection effect is achieved. Efficient arrangement and improvement of rat disinfection tasks.
Patent Information
- Application Number
- CN202411557892.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-04
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing rat pest prevention and control methods rely on manual patrol and fixed modes, which are difficult to adapt to the dynamic changes in rat activities, resulting in poor resource waste and disinfection effects.
Using an intelligent disinfection operation and maintenance management system, data preprocessing, clustering, future prediction, hazard rating and path analysis are carried out by collecting mouse activity data, using clustering analysis, Markov prediction, ant colony algorithm and genetic algorithm. Realize the dynamic arrangement of rat disinfection tasks.
It improves the efficiency and effectiveness of disinfection work, ensures efficient utilization of resources, realizes rapid prediction and analysis of mouse behavior patterns, and ensures timely processing of key areas.
Smart Images

Figure CN119067455A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of rodent pest control, and more specifically, to an intelligent pest control and maintenance management system and method. Background Art
[0002] Rodents are one of the most destructive pests in the world and are widely distributed in urban and rural areas, threatening food safety, public health, and the safe operation of infrastructure. Rodent pest control has become a routine and ongoing problem.
[0003] Most of the existing rodent control methods rely on manual inspections, deployment and treatment, which are cumbersome and time-consuming. Due to the influence of human factors, the operator's work ability, experience level and understanding of the on-site environment directly affect the control effect, making it difficult for the control work to achieve the expected high-efficiency standards; most traditional pest control equipment uses preset modes to work, such as timed release of rat poison or fixed-frequency ultrasonic rodent repellent; such fixed modes cannot adapt to the dynamic changes in rodent activities, especially in environments where the frequency of rodent activities fluctuates greatly. This "fixed pest control" method leads to waste of resources, or it is difficult to achieve the best pest control effect.
[0004] In view of this, the present invention proposes an intelligent disinfection operation and maintenance management system and method to solve the above problems. Summary of the invention
[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent disinfection operation and maintenance management method, comprising: S1. Collect activity data of rodents and pre-process the activity data to obtain effective activity data; S2. Use cluster analysis algorithm to perform cluster analysis on effective activity data and obtain Regional clustering; S3, based on the Markov prediction algorithm, the regional clusters are predicted to obtain the future probability map, and the ant colony algorithm is used to rate the danger of the future probability map to obtain the future risk map; S4. Use genetic algorithms to perform path analysis on future risk maps to obtain the best extermination path; arrange rodent extermination tasks based on the best extermination path to achieve rapid extermination of rodents.
[0006] Furthermore, the activity data includes: activity frequency, activity trajectory, activity time and area mark.
[0007] Furthermore, the method of preprocessing the activity data includes: A low-pass filter and a cutoff frequency of the low-pass filter are preset, and the activity data is input into the set low-pass filter. The low-pass filter separates the frequency components of the activity frequency and the activity trajectory, and filters out high-frequency noise higher than the cutoff frequency to obtain filtered activity data. An anomaly detection algorithm is used to identify anomalies in the filtered activity data, and the filtered activity data identified as abnormal is marked as abnormal, and the filtered activity data marked as abnormal in the filtered activity data are eliminated, and the remaining filtered activity data constitutes valid activity data.
[0008] Furthermore, the method of clustering the effective activity data includes: Preset Category Collection , preset the stop mark, initialize the stop mark to an infinite value, preset the best group, initialize the best group to an empty value; traverse the classification set, based on the traversed classification value and the activity frequency in the valid activity data, taking the first valid activity data as the starting point, and The value of the modulus calculation is used as the selection span to select the initial cluster center until the number of selected initial cluster centers is Stop at 1:00 and get Initial cluster centers; Based on the initial cluster centers, the Euclidean distance formula is used to calculate the distance between each data point in the valid activity data. The distance between the activity frequency in and the initial cluster center, and the initial cluster center corresponding to the minimum distance is selected as The clustering points of Assign to cluster points and obtain Data clusters are formed, and the average value of the activity frequency of the data in each data cluster is calculated. The average value is used as the new value of the initial cluster center. This is repeated until the value of the initial cluster center no longer changes. The data cluster with the smallest value of the initial cluster center is selected as the marked cluster. The sum of the squares of the distances between the activity frequency of the data in the marked cluster and the initial cluster center corresponding to the marked cluster is calculated as the minimum cluster distance. When the value of the minimum cluster distance is less than or equal to the value of the stop mark, the value of the minimum cluster distance is used as the new value of the stop mark, and the current data cluster is used as the best grouping. When the value of the minimum cluster distance is greater than the value of the stop mark, the traversal is stopped, and the output best grouping is the regional clustering.
[0009] Furthermore, the method of performing future prediction on regional clustering includes: Based on regional cluster definition cluster states, and the value of each cluster state is the average of the activity frequencies of the data in the corresponding regional cluster. Based on the cluster state, a The probability vector of the initial state is the ratio of the cluster state value to the sum of all cluster state values. Different regional labels are based on the cluster state labels defined by regional clustering, and the regional labeling state diagram is obtained. Based on the regional labeling state diagram, the statistical estimation algorithm is used to calculate the transition probability between different clustering states, and the clustering state transition matrix is constructed based on the transition probability. When a type of state transition has no corresponding transition probability, zero filling is performed; based on the initial state probability vector and the cluster state transfer matrix, the future cluster state is probabilistically predicted to obtain the future state probability vector; based on the predicted probability in the future probability vector, the regional markers are probabilistically marked to obtain the future probability map.
[0010] Furthermore, the method for obtaining the transition probability includes: Based on the activity trajectory data, the transition situations between all regional markers are counted to obtain the regional transfer set. The regional transfer set is clustered and mapped through the regional marker state diagram to obtain the state transfer set. The total number of each state transfer category in the state transfer set is counted. Based on the total number of state transfer categories and the total amount of data in the state transfer set, the transition probability between all cluster states is calculated. The value of the transition probability is the ratio of the total number of state transfer categories to the total amount of data in the state transfer set.
[0011] Furthermore, the method of performing risk rating on the future probability map includes: Initialize E ants, assign an initial pheromone concentration to each ant, and randomly assign each ant to a different area mark. Initialize the assigned concentration , based on the probabilities marked at different area markers in the future probability map, assign the initial pheromone concentration at each area marker; The size of the initial pheromone concentration is the product of the probability of marking the area mark and the distribution concentration. Each ant moves between the area marks and selects the next location to move to based on the initial pheromone concentration at the area mark. The ant moves to the area mark with high pheromone concentration. After the area mark of each ant changes, the pheromone concentration at the area mark of the ant is updated. When the initial pheromone concentration of the ant is zero, the ant will stop moving. When all ants stop moving, the area mark is rated for danger based on the ratio of the number of ants at the area mark to the total number of ants, and a future risk map is obtained; the danger rating is a specific value, and the size of the value is the ratio of the number of ants at the area mark to the total number of ants.
[0012] Furthermore, the method of performing path analysis on the future risk map includes: A permutation and combination algorithm is used to calculate the combination of all pairs of regional markers to obtain regional marker combinations. All regional marker combinations constitute a regional marker combination set. The regional marker combination set is traversed, and one regional marker in the regional marker combination is used as the starting point and the other regional marker is used as the ending point. The starting point is used as the start point and the ending point is used as the termination condition. The path is explored through a recursive algorithm to obtain the local optimal path of each regional marker combination; each local optimal path is represented as a chromosome, and the genes in the chromosome correspond to the regional markers in the local optimal path; a fitness formula is defined; for each chromosome, the fitness of each chromosome is calculated according to the fitness formula, and the parent chromosome is selected through the tournament selection algorithm according to the fitness value, and the parent chromosome with the largest fitness value among the parent chromosomes is selected. as pre-existing chromosomes; pair the parent chromosomes in pairs to obtain exchange combinations, preset the Uth gene as the exchange node, exchange the genes of the parent chromosome after the exchange node in each exchange combination to generate daughter chromosomes; through the random mutation algorithm, randomly select a gene of the daughter chromosome to replace the gene at the corresponding position of the pre-existing chromosome to obtain a mutant chromosome; calculate the fitness of the mutant chromosome according to the fitness formula, compare the mutant chromosome with the pre-existing chromosome one by one, when the fitness value of the mutant chromosome is greater than the fitness value of the pre-existing chromosome and the number of genes of the mutant chromosome is less than the number of genes of the pre-existing chromosome, replace the pre-existing chromosome with the mutant chromosome as the new pre-existing chromosome, repeat until the fitness value of the pre-existing chromosome no longer changes, and the pre-existing chromosome output at this time is the optimal disinfecting path.
[0013] Furthermore, the path exploration method includes: A preset path set is used, and the exploration starts from the starting point. The starting point is counted into the path set, and the starting point is marked as "yes". The moving mark includes "yes" and "no". "Yes" means that the area mark has been explored, and "no" means that the area mark has not been explored. The area mark with the highest danger rating and a moving mark of "no" among the area marks adjacent to the starting point is selected as the exploration point, and the exploration point is used as the next moving position. When the exploration point is reached, the exploration point is counted into the path set, and the exploration point is marked as "yes". When the moving mark value of all area marks adjacent to the exploration point is "yes", the exploration point is removed from the path set and returns to the previous exploration point to continue exploring. Repeat until the termination condition is met. At this time, the output path set is the local optimal path.
[0014] The present invention provides an intelligent disinfection operation and maintenance management system, comprising: Data collection module: collects the activity data of rodents and pre-processes the activity data to obtain effective activity data; Data analysis module: Use cluster analysis algorithm to perform cluster analysis on effective activity data to obtain Regional clustering; Risk prediction module: includes prediction unit and rating unit. The prediction unit uses Markov prediction algorithm to make future predictions on regional clusters to obtain future probability maps. The rating unit uses ant colony algorithm to rate the future probability maps and obtain future risk maps. Solution generation module: Use genetic algorithms to analyze the path of future risk maps to obtain the best extermination path; arrange rodent extermination tasks based on the best extermination path to achieve rapid extermination of rodents.
[0015] The technical effects and advantages of the intelligent pest control and maintenance management system and method of the present invention are as follows: The present invention removes high-frequency noise through a low-pass filtering algorithm to ensure the accuracy of the data. At the same time, the anomaly detection algorithm further improves the reliability of the data, ensures the effectiveness of subsequent analysis, and provides a solid foundation for intelligent disinfection. By adopting a cluster analysis algorithm, complex effective activity data is simplified into regional clusters, making the data structure clearer and easier to understand, so that the system can predict and analyze rodent behavior patterns more quickly, improving the efficiency of operation and maintenance management. The ant colony algorithm is used to derive the danger rating of different areas, which provides clear guidance for subsequent disinfection path planning and ensures that key areas of disinfection work are dealt with in a timely manner. The genetic algorithm is used to find the best disinfection path, and the best disinfection path can cover the highest risk area with minimal resource consumption, thereby achieving efficient disinfection task arrangement. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 A schematic diagram of an intelligent disinfection operation and maintenance management method of the present invention; Figure 2 It is a schematic diagram of an intelligent disinfection operation and maintenance management system of the present invention. DETAILED DESCRIPTION
[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0018] Example 1 See also Figure 1 As shown, a flow chart of an intelligent pest control and operation and maintenance management method of this embodiment specifically includes: S1, collecting activity data of rodents, and pre-processing the activity data to obtain effective activity data.
[0019] Specifically, the activity data includes activity frequency, activity trajectory, activity time and regional mark. The regional mark is the mark information of an area. The rodent control area is grid-divided by the area bisection algorithm to obtain a grid map. Each grid in the grid map is marked with a different serial number to distinguish different grids. The serial numbers of different grids are the mark information of an area. The activity trajectory is the activity route of rodents in the entire rodent control area. It is numerical data. The data content includes time points and regional marks. The activity time is the specific time period of rodent activities. The activity frequency is the number of times the rodent moves in the grid corresponding to the regional mark. For example: { [T 0 , P 0 , M 0 ], [T 1 , P 1 , M 1 ],...,[ T n , P n , M n ]} , which is the representation of activity data, where [T 0 , P 0 , M 0 ] Indicates at time When the rodent is in the area marked as area, and the activity frequency is ; [T 1 , P 1 , M 1 ] Indicates at time When the rodent is in the area marked as area, and the activity frequency is ; [T n , P n , M n ] Indicates at time When the rodent is in the area marked as area, and the activity frequency is ; to The time period represents the activity time; the activity trajectory is obtained through the ultrasonic sensor, which senses the movement of mice by emitting ultrasonic waves and detecting reflected signals, and can capture the position information of mice more accurately. The activity frequency is obtained through the acceleration sensor, which can detect the subtle movements of mice more accurately.
[0020] Specifically, the methods for preprocessing activity data include: The high-frequency noise in the activity data is removed by the low-pass filtering algorithm. The type of low-pass filter is set. Commonly used low-pass filters include Buster filter and Chebyshev filter. According to the characteristics of the sensor data, the cut-off frequency is preset to filter out the high-frequency noise. For example, the noise of ultrasonic sensors and acceleration sensors are mainly concentrated in the high-frequency segment, so the cut-off frequency can be selected as 9HZ, 10HZ or 11HZ; the activity data is input into the set low-pass filter, which separates the frequency components of the activity frequency and the activity trajectory, and filters out the high-frequency noise above the cut-off frequency to obtain the filtered data. Activity data; the activity data processed by the low-pass filtering algorithm is smoother, avoiding the influence of short-term noise caused by electromagnetic interference or sensor hardware, and ensuring the accuracy of rodent activity data; the abnormality detection algorithm is used to identify abnormalities in the filtered activity data, and the filtered activity data identified as abnormal are marked as abnormal, and the filtered activity data marked as abnormal are removed from the filtered activity data, and the remaining filtered activity data constitute valid activity data; removing the filtered activity data marked as abnormal can more accurately reflect the overall trend and pattern of the data, thereby obtaining more reliable analysis results, making the analysis results more neutral and fair; Commonly used anomaly detection algorithms include box plot method, standard score method and median absolute deviation method. Taking the activity frequency in the filtered activity data as an example, the method of marking the data in the activity frequency by the box plot method includes: counting the quantile indicators in the activity frequency, and the quantile indicators include: minimum value , median and maximum value , define the first quantile based on the quantile index and the second quantile , the value of the first quantile is the average of the sum of the minimum value and the median, and the value of the second quantile is the average of the sum of the maximum value and the median. The upper and lower boundaries of the activity frequency are defined based on the first and second quantiles. The calculation formula for the upper boundary is: , the calculation formula of the lower boundary is: ,in, It stands for quantile difference. The value of quantile difference is the difference between the second quantile and the first quantile. Data with a frequency greater than the upper boundary and data with a frequency less than the lower boundary are marked as abnormal.
[0021] S2. Use cluster analysis algorithm to perform cluster analysis on effective activity data and obtain Regional clustering; Specifically, the methods of cluster analysis include: Preset Category Collection , ; Preset a stop mark, initialize the stop mark to an infinite value, preset the best group, initialize the best group to a null value; where, Represents the total amount of valid activity data, traversing the classification set, based on the traversed classification value and the activity frequency in the valid activity data, select The activity frequencies are used as the initial cluster centers. The selection methods of the initial cluster centers include: using the first valid activity data as the starting point, and The value of the modulus calculation is used as the selection span to select the initial cluster center until the number of selected initial cluster centers is Stop at 1:00 and get Initial cluster centers; Based on the initial cluster center, the Euclidean distance formula is used to calculate the distance between each data in the effective activity data. The distance between the activity frequency in and the initial cluster center, and the initial cluster center corresponding to the minimum distance is selected as The clustering points of Assign to cluster points and obtain Data clusters are formed, and the average value of the activity frequency of the data in each data cluster is calculated. The average value is used as the new value of the initial cluster center. This is repeated until the value of the initial cluster center no longer changes. The data cluster with the smallest value of the initial cluster center is selected as the marked cluster. The sum of the squares of the distances between the activity frequency of the data in the marked cluster and the initial cluster center corresponding to the marked cluster is calculated as the minimum cluster distance. When the value of the minimum cluster distance is less than or equal to the value of the stop mark, the value of the minimum cluster distance is used as the new value of the stop mark, and the current data cluster is used as the best grouping. When the value of the minimum cluster distance is greater than the value of the stop mark, the traversal is stopped, and the output best grouping is the regional clustering. The effective activity data set is simplified through cluster analysis, making the structure of the effective activity data clearer and easier to understand, which is convenient for subsequent analysis and processing of the effective activity data.
[0022] S3, based on the Markov prediction algorithm, the regional clusters are predicted to obtain the future probability map, and the ant colony algorithm is used to rate the danger of the future probability map to obtain the future risk map; Specifically, the methods for making future predictions include: Based on regional cluster definition clustering states, where The value of is the total number of clusters in the regional clustering, and the value of each clustering state is the average value of the activity frequency of the data in the corresponding regional clustering. Based on the clustering state, a The initial state probability vector X=[ O 1 ,..., O m ] ,in, represents the probability of the first clustering state, Represents the probability of the mth cluster state, and the value of m is The values of are consistent in size, and the probability of each cluster state is the ratio of the cluster state value to the sum of all cluster state values. The larger the value of the cluster state, the more active the mice are in the area corresponding to the area mark in the area cluster, the stronger the attraction of the area to the mice, and the greater the probability of the mice moving to the area. For different area marks, the corresponding cluster states on the cluster state marks defined based on the area cluster are obtained to obtain the area mark state diagram, and the statistical estimation algorithm is used based on the area mark state diagram to calculate the transition probability between different cluster states. The method for obtaining the transition probability includes: based on the activity trajectory data, the transition situation between all area marks is counted to obtain the area transfer set, the area transfer set is clustered by the area mark state diagram, and the state transfer set is obtained. The total number of state transfer categories in the state transfer set is counted, and the transition probability between all cluster states is calculated based on the total number of state transfer categories and the total amount of data in the state transfer set. The value of the transition probability is the ratio of the total number of state transfer categories to the total amount of data in the state transfer set, and the cluster state transfer matrix is constructed based on the transition probability. , ;in, Represents the clustering state Transition to clustering state The probability of Clustering status Transition to clustering state The probability of Represents the clustering state Transition to clustering state The probability of Represents the clustering state Transition to clustering state The probability of represents the first clustering state, Representative Cluster states, when a state transition has no corresponding transition probability, it is filled with zeros; based on the initial state probability vector and the cluster state transition matrix, the future cluster state is predicted probabilistically. The calculation formula for probability prediction is: ;in Representatives passed The probability vector of future states at time steps, Represents the cluster state transition matrix power; based on the predicted probability in the future probability vector, the regional markers are probabilistically marked to obtain a future probability map; the probability-driven prediction can remain robust in uncertainty and noisy data, and the randomness of rodent behavior can be intuitively represented with specific data, making it easier to gain insight into the laws of rodent behavior.
[0023] For example: Taking the activity track: { [T 0 , P 0 , M 0 ], [T 1 , P 1 , M 1 ],[ T 2 , P 2 , M 2 ],[ T 3 , P 3 , M 3 ],[ T 4 , P 4 , M 4 ]} For example, the region transfer set is: { [M 0 → M 1 ], [M 1 → M 2 ], [M 2 → M 3 ], [M 3 → M 4 ]} , assuming that the region marking state diagram is: {[ M 0 , Ty 1 ],[ M 1 , Ty 2 ],[ M 2 , Ty 1 ],[ M 3 , Ty 2 ],[ M 4 , Ty 3 ]} ,in , and Represents different clustering states, then the state transition set is: { [Ty 1 → Ty 2 ], [Ty 2 → Ty 1 ], [Ty 1 → Ty 2 ], [Ty 2 → Ty 3 ]} , based on the state transition set statistics, the total state transition number is the size of the state transition set, and the number of each state transition category is the total number of the same state transition category; the total state transition number is 4, and the state transition is [Ty 1 → Ty 2 ] The number of is 2, and the same [Ty 2 → Ty 1 ] The number of is 1, [Ty 2 → Ty 3 ] The number of is 1, and the ratio of the number of each state transition category to the total number of state transitions is used as the probability of transition between different clustering states. [Ty 1 → Ty 2 ] The probability of , [Ty 2 → Ty 1 ] The probability of , [Ty 2 → Ty 3 ] The probability of , based on the probability of transition between different clustering states, the clustering state transition matrix is constructed, then the clustering state transition matrix is .
[0024] Specifically, the methods for risk rating include: Initialize E ants, where the value of E is greater than the total number of regional markers, assign an initial pheromone concentration to each ant, and randomly assign each ant to a different regional marker to initialize the assigned concentration. , is a minimum value greater than zero. Based on the probability of marking at different regional markers in the future probability map, the initial pheromone concentration at each regional marker is assigned; the size of the initial pheromone concentration is the product of the probability of marking at the regional marker and the assigned concentration. Each ant moves between regional markers and selects the next location to move to based on the initial pheromone concentration at the regional marker; the ant will move to the regional marker with high pheromone concentration; after each ant's regional marker changes, the pheromone concentration at the regional marker where the ant is located is updated. When the initial pheromone concentration of the ant is zero, the ant will stop moving. When all ants stop moving, the regional marker is rated for danger based on the ratio of the number of ants at the regional marker to the total number of ants, and a future risk map is obtained; the danger rating is a specific value, and the size of the value is the ratio of the number of ants at the regional marker to the total number of ants. The larger the value of the danger rating, the more dangerous the regional marker is; the formula for updating the pheromone concentration at the regional marker is: ;in, Representative The updated value of the pheromone concentration at the region marker, Represents the pheromone volatility coefficient, which is used to simulate the normal volatility of pheromone concentration. Representative The original value of the pheromone concentration at the area marker, Representative The probability of the region marker, The initial pheromone concentration assigned to the representative ants.
[0025] S4. Use genetic algorithms to analyze the path of future risk maps and obtain the best extermination path; arrange rodent extermination tasks based on the best extermination path to achieve rapid extermination of rodents; Specifically, the methods of using genetic algorithm future risk graphs for path analysis include: The permutation and combination algorithm is used to calculate the combination of all the regional markers in pairs, and the regional marker combination is obtained. All regional marker combinations constitute a regional marker combination set. The regional marker combination set is traversed, and one regional marker in the regional marker combination is used as the starting point and the other regional marker is used as the ending point. The starting point is used as the start point, and the arrival at the ending point is used as the termination condition. The path exploration is performed through the recursive algorithm to obtain the local optimal path of each regional marker combination. The method of performing path exploration includes: presetting a path set, starting exploration from a starting point, counting the starting point into the path set, and moving marking the starting point as "yes", the moving marking includes "yes" and "no", "yes" represents that the area mark has been explored, and "no" represents that the area mark has not been explored, selecting the area mark with the highest danger rating and the moving mark of "no" among the area marks adjacent to the starting point as the exploration point, and taking the exploration point as the next moving position, when reaching the exploration point position, counting the exploration point into the path set, and moving marking the exploration point as "yes", when the moving mark values of all area marks adjacent to the exploration point are "yes", removing the exploration point from the path set and returning to the previous exploration point to continue exploring, repeating until the termination condition is reached, and the output path set is the local optimal path; Each local optimal path is represented as a chromosome, and the genes in the chromosome correspond to the regional labels in the local optimal path; define the fitness formula ;in, represents fitness, Representative The risk rating of each gene, Represents the total number of genes in the chromosome; for each chromosome, the fitness of each chromosome is calculated according to the fitness formula, and the parent chromosome is selected through the tournament selection algorithm according to the fitness value, and the parent chromosome with the largest fitness value among the parent chromosomes is selected as the pre-stored chromosome; the parent chromosomes are paired in pairs to obtain exchange combinations, and the Uth gene is preset as the exchange node. The genes of the parent chromosome after the exchange node in each exchange combination are exchanged to generate daughter chromosomes; through the random mutation algorithm, a gene of the daughter chromosome is randomly selected to replace the gene at the corresponding position of the pre-stored chromosome to obtain a mutant chromosome; the mutant chromosome is calculated according to the fitness formula The fitness of the chromosome is determined by comparing the mutated chromosome with the pre-existing chromosome one by one. When the fitness value of the mutated chromosome is greater than the fitness value of the pre-existing chromosome and the number of genes of the mutated chromosome is less than the number of genes of the pre-existing chromosome, the mutated chromosome replaces the pre-existing chromosome as the new pre-existing chromosome, and the process is repeated until the fitness value of the pre-existing chromosome no longer changes. At this time, the output pre-existing chromosome is the optimal disinfecting path. The optimal killing path is the path with the largest sum of the hazard ratings of regional markers under the condition of containing the least regional markers. The rodent disinfecting task is arranged based on the optimal disinfecting path to achieve rapid rodent disinfecting and obtain the maximum disinfecting benefit with the least resource consumption.
[0026] This embodiment uses a low-pass filtering algorithm to remove high-frequency noise to ensure the accuracy of the data. At the same time, the anomaly detection algorithm further improves the reliability of the data, ensures the effectiveness of subsequent analysis, and provides a solid foundation for intelligent disinfection. By adopting a cluster analysis algorithm, complex effective activity data is simplified into regional clusters, making the data structure clearer and easier to understand, so that the system can predict and analyze rodent behavior patterns more quickly, improving the efficiency of operation and maintenance management. The ant colony algorithm is used to derive the danger rating of different areas, which provides clear guidance for subsequent disinfection path planning and ensures that key areas of disinfection work are dealt with in a timely manner. The genetic algorithm is used to find the best disinfection path, and the best disinfection path can cover the highest risk area with minimal resource consumption, thereby achieving efficient disinfection task arrangement.
[0027] Example 2 See also Figure 2 As shown, the part not described in detail in this embodiment is described in Example 1, which provides an intelligent disinfection operation and maintenance management system, including: Data collection module: collects the activity data of rodents and pre-processes the activity data to obtain effective activity data; Data analysis module: Use cluster analysis algorithm to perform cluster analysis on effective activity data to obtain Regional clustering; Risk prediction module: includes prediction unit and rating unit. The prediction unit uses Markov prediction algorithm to make future predictions on regional clusters to obtain future probability maps. The rating unit uses ant colony algorithm to rate the future probability maps and obtain future risk maps. Solution generation module: Use genetic algorithm to analyze the path of future risk map and obtain the best disinfecting path; arrange rodent disinfecting tasks based on the best disinfecting path to achieve rapid disinfecting of rodents; The modules are connected to each other via wired and / or wireless means to achieve data transmission between modules.
[0028] Example 3 This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the intelligent disinfection and operation and maintenance management method provided above is implemented.
[0029] Since the electronic device introduced in this embodiment is an electronic device used to implement an intelligent disinfection operation and maintenance management method in the embodiment of this application, based on the intelligent disinfection operation and maintenance management method introduced in the embodiment of this application, the technical personnel of this field can understand the specific implementation method of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application is not introduced in detail here. As long as the technical personnel of this field implement the electronic device used in the intelligent disinfection operation and maintenance management method in the embodiment of this application, it belongs to the scope of protection of this application.
[0030] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.
[0031] The above are only preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.
Claims
1. An intelligent pest control and maintenance management method, characterized in that: include: S1. Collecting activity data of rodents and preprocessing the activity data to obtain effective activity data; S2. Use cluster analysis algorithm to perform cluster analysis on effective activity data and obtain Regional clustering; S3, based on the Markov prediction algorithm, the regional clusters are predicted to obtain the future probability map, and the ant colony algorithm is used to rate the danger of the future probability map to obtain the future risk map; S4. Use genetic algorithms to perform path analysis on future risk maps to obtain the best extermination path; arrange rodent extermination tasks based on the best extermination path to achieve rapid extermination of rodents.
2. The intelligent disinfection operation and maintenance management method according to claim 1 is characterized in that: The activity data includes: activity frequency, activity track, activity time and area mark.
3. The intelligent disinfection operation and maintenance management method according to claim 2 is characterized in that: The method of preprocessing the activity data includes: A low-pass filter and a cutoff frequency of the low-pass filter are preset, and the activity data is input into the set low-pass filter. The low-pass filter separates the frequency components of the activity frequency and the activity trajectory, and filters out high-frequency noise higher than the cutoff frequency to obtain filtered activity data. An anomaly detection algorithm is used to identify anomalies in the filtered activity data, and the filtered activity data identified as abnormal is marked as abnormal, and the filtered activity data marked as abnormal in the filtered activity data are eliminated, and the remaining filtered activity data constitutes valid activity data.
4. The intelligent disinfection operation and maintenance management method according to claim 3 is characterized in that: The method of clustering the effective activity data includes: Preset Category Collection , preset the stop mark, initialize the stop mark to an infinite value, preset the best group, initialize the best group to an empty value; traverse the classification set, based on the traversed classification value and the activity frequency in the valid activity data, taking the first valid activity data as the starting point, and The value of the modulus calculation is used as the selection span to select the initial cluster center until the number of selected initial cluster centers is Stop at 1:00 and get Initial cluster centers; Based on the initial cluster centers, the Euclidean distance formula is used to calculate the distance between each data point in the valid activity data. The distance between the activity frequency in and the initial cluster center, and the initial cluster center corresponding to the minimum distance is selected as The clustering points of Assign to cluster points and obtain Data clusters are formed, and the average value of the activity frequency of the data in each data cluster is calculated. The average value is used as the new value of the initial cluster center. This is repeated until the value of the initial cluster center no longer changes. The data cluster with the smallest value of the initial cluster center is selected as the marked cluster. The sum of the squares of the distances between the activity frequency of the data in the marked cluster and the initial cluster center corresponding to the marked cluster is calculated as the minimum cluster distance. When the value of the minimum cluster distance is less than or equal to the value of the stop mark, the value of the minimum cluster distance is used as the new value of the stop mark, and the current data cluster is used as the best grouping. When the value of the minimum cluster distance is greater than the value of the stop mark, the traversal is stopped, and the output best grouping is the regional clustering.
5. The intelligent disinfection operation and maintenance management method according to claim 4 is characterized in that: The method for making future predictions for regional clusters includes: Based on regional cluster definition cluster states, and the value of each cluster state is the average of the activity frequencies of the data in the corresponding regional cluster. Based on the cluster state, a The probability vector of the initial state is the ratio of the cluster state value to the sum of all cluster state values. Different regional labels are based on the cluster state labels defined by regional clustering, and the regional labeling state diagram is obtained. Based on the regional labeling state diagram, the statistical estimation algorithm is used to calculate the transition probability between different clustering states, and the clustering state transition matrix is constructed based on the transition probability. When a type of state transition has no corresponding transition probability, zero filling is performed; based on the initial state probability vector and the cluster state transfer matrix, the future cluster state is probabilistically predicted to obtain the future state probability vector; based on the predicted probability in the future probability vector, the regional markers are probabilistically marked to obtain the future probability map.
6. The intelligent disinfection operation and maintenance management method according to claim 5 is characterized in that: The method for obtaining the transition probability includes: Based on the activity trajectory data, the transition situations between all regional markers are counted to obtain the regional transfer set. The regional transfer set is clustered and mapped through the regional marker state diagram to obtain the state transfer set. The total number of each state transfer category in the state transfer set is counted. Based on the total number of state transfer categories and the total amount of data in the state transfer set, the transition probability between all cluster states is calculated. The value of the transition probability is the ratio of the total number of state transfer categories to the total amount of data in the state transfer set.
7. The intelligent disinfection operation and maintenance management method according to claim 6 is characterized in that: The method of performing risk rating on the future probability map includes: Initialize E ants, assign an initial pheromone concentration to each ant, and randomly assign each ant to a different area mark. Initialize the assigned concentration , based on the probabilities marked at different area markers in the future probability map, assign the initial pheromone concentration at each area marker; The size of the initial pheromone concentration is the product of the probability of marking the area mark and the distribution concentration. Each ant moves between the area marks and selects the next location to move to based on the initial pheromone concentration at the area mark. The ant moves to the area mark with high pheromone concentration. After the area mark of each ant changes, the pheromone concentration at the area mark of the ant is updated. When the initial pheromone concentration of the ant is zero, the ant will stop moving. When all ants stop moving, the area mark is rated for danger based on the ratio of the number of ants at the area mark to the total number of ants, and a future risk map is obtained; the danger rating is a specific value, and the size of the value is the ratio of the number of ants at the area mark to the total number of ants.
8. The intelligent disinfection operation and maintenance management method according to claim 7 is characterized in that: The method of performing path analysis on the future risk map includes: A permutation and combination algorithm is used to calculate the combination of all pairs of regional markers to obtain regional marker combinations. All regional marker combinations constitute a regional marker combination set. The regional marker combination set is traversed, and one regional marker in the regional marker combination is used as the starting point and the other regional marker is used as the ending point. The starting point is used as the start point and the ending point is used as the termination condition. The path is explored through a recursive algorithm to obtain the local optimal path of each regional marker combination; each local optimal path is represented as a chromosome, and the genes in the chromosome correspond to the regional markers in the local optimal path; a fitness formula is defined; for each chromosome, the fitness of each chromosome is calculated according to the fitness formula, and the parent chromosome is selected through the tournament selection algorithm according to the fitness value, and the parent chromosome with the largest fitness value among the parent chromosomes is selected. as pre-existing chromosomes; pair the parent chromosomes in pairs to obtain exchange combinations, preset the Uth gene as the exchange node, exchange the genes of the parent chromosome after the exchange node in each exchange combination to generate daughter chromosomes; through the random mutation algorithm, randomly select a gene of the daughter chromosome to replace the gene at the corresponding position of the pre-existing chromosome to obtain a mutant chromosome; calculate the fitness of the mutant chromosome according to the fitness formula, compare the mutant chromosome with the pre-existing chromosome one by one, when the fitness value of the mutant chromosome is greater than the fitness value of the pre-existing chromosome and the number of genes of the mutant chromosome is less than the number of genes of the pre-existing chromosome, replace the pre-existing chromosome with the mutant chromosome as the new pre-existing chromosome, repeat until the fitness value of the pre-existing chromosome no longer changes, and the pre-existing chromosome output at this time is the optimal disinfecting path.
9. The intelligent disinfection operation and maintenance management method according to claim 8 is characterized in that: The method of performing path exploration includes: A preset path set is used, and the exploration starts from the starting point. The starting point is counted into the path set, and the starting point is marked as "yes". The moving mark includes "yes" and "no". "Yes" means that the area mark has been explored, and "no" means that the area mark has not been explored. The area mark with the highest danger rating and a moving mark of "no" among the area marks adjacent to the starting point is selected as the exploration point, and the exploration point is used as the next moving position. When the exploration point is reached, the exploration point is counted into the path set, and the exploration point is marked as "yes". When the moving mark value of all area marks adjacent to the exploration point is "yes", the exploration point is removed from the path set and returns to the previous exploration point to continue exploring. Repeat until the termination condition is met. At this time, the output path set is the local optimal path.
10. An intelligent pest control and operation management system, which is used to implement an intelligent pest control and operation management method according to any one of claims 1 to 9, characterized in that: include: Data collection module: collects the activity data of rodents and pre-processes the activity data to obtain effective activity data; Data analysis module: Use cluster analysis algorithm to perform cluster analysis on effective activity data to obtain Regional clustering; Risk prediction module: includes prediction unit and rating unit. The prediction unit uses Markov prediction algorithm to make future predictions on regional clusters to obtain future probability maps. The rating unit uses ant colony algorithm to rate the future probability maps and obtain future risk maps. Solution generation module: Use genetic algorithms to analyze the path of future risk maps to obtain the best extermination path; arrange rodent extermination tasks based on the best extermination path to achieve rapid extermination of rodents.
Citation Information
Patent Citations
Intelligent disinfection system and method
CN113144264A
Forest fire early warning method, system and equipment based on satellite Internet of Things
CN118887767A
PLS truck system with roller system for loading and unloading
KR102588917B1
