Smart park integrated management system based on intelligent acquisition

Abstract

The invention discloses a smart park comprehensive management system based on intelligent collection, and relates to the technical field of smart park comprehensive management, and the system comprises a collection and generation module which collects original data, segments a service area through a Voronoi graph, and generates a shortest distance matrix in combination with a Floyd algorithm; the calculation and iteration module is used for calculating facility priority scores, generating a facility start-stop sequence and scheduling individuals in combination with the shortest distance matrix, and carrying out iterative optimization to obtain an initial solution; the optimization module is used for generating an elite and a common population based on the initial solution to perform initialization, and performing optimization through disturbance, path selection probability and a pheromone diffusion mechanism in combination with a Hamming distance difference evaluation and cross replacement strategy to obtain an optimal solution; according to the invention, the position of the facility can be accurately adjusted, the facility arrangement of the park is more reasonable, and the service range is maximized, so that the overall service quality and the operation efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of integrated management of smart parks, and in particular to an integrated management system of smart parks based on intelligent collection. Background Art

[0002] In recent years, smart parks, as a composite management system integrating the Internet of Things, big data analysis, and artificial intelligence technologies, have gradually become an important part of modern urban infrastructure. Smart parks use intelligent data collection from sensors and equipment to monitor the operating status and environmental data of various facilities in the park in real time, and then carry out intelligent management and optimization. Especially in the field of energy management and resource scheduling, intelligent data collection and decision-making systems have greatly improved the operating efficiency and energy-saving effects of the park. With the popularization of Internet of Things technology, more and more parks have begun to use sensor data collection and analysis methods to carry out dynamic monitoring and optimized scheduling.

[0003] However, most traditional smart park comprehensive management systems lack accurate facility location optimization methods, making it difficult to allocate resources reasonably, resulting in unsatisfactory facility coverage and service quality. In addition, traditional smart park comprehensive management systems often rely on static rules to determine the start and stop sequence of equipment, and are unable to respond to environmental changes in real time or perform intelligent optimization based on dynamic service needs. Moreover, equipment scheduling and load distribution mostly rely on manually set thresholds and simple start and stop controls, resulting in waste of park energy efficiency or excessive equipment load. Summary of the Invention

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

[0005] Therefore, the present invention provides a smart park comprehensive management system based on intelligent data collection, which solves the problems that traditional smart park comprehensive management systems are difficult to reasonably allocate resources and cannot respond to environmental changes in real time or perform intelligent optimization according to dynamic service needs, and that equipment scheduling and load distribution mostly rely on manually set thresholds and simple start-stop control.

[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 smart park integrated management system based on intelligent collection, which includes:

[0008] The acquisition and generation module collects raw data, segments the service area using the Voronoi diagram, and combines the Floyd algorithm to generate the shortest distance matrix;

[0009] The original data includes the facility status of the lighting and ventilation facilities in the park and the park environment data;

[0010] The calculation and iteration module calculates the facility priority scores, generates the facility start and stop sequence based on the shortest distance matrix, and performs iterative optimization on the scheduling individuals to obtain the initial solution;

[0011] The initial solution includes a start-stop sequence and a power setting combination;

[0012] The optimization module generates elite and common populations based on the initial solution for initialization, and optimizes through perturbation, path selection probability, and pheromone diffusion mechanism, combined with Hamming distance difference evaluation and cross-replacement strategy to obtain the optimal solution;

[0013] The optimal solution includes an optimized start-stop sequence and power setting combination;

[0014] The early warning module builds an evaluation model to predict the status of facilities, triggers optimal solution recalculation and early warning based on the prediction results, and implements corresponding measures;

[0015] The data management module displays the data through a visual interface, uses the database to store the data, and then encrypts it using an encryption algorithm.

[0016] As a preferred solution of the intelligent park integrated management system based on intelligent data collection described in the present invention, the steps of collecting raw data, segmenting the service area using the Voronoi diagram, and generating the shortest distance matrix in combination with the Floyd algorithm include the following steps:

[0017] Use IoT sensors to collect real-time data on the status of lighting and ventilation facilities within the park and on the park environment;

[0018] The facility status data includes the switch status, energy consumption and location of lighting and ventilation facilities;

[0019] The environmental data includes temperature and light intensity;

[0020] The park is divided into sections using the grid division method. After setting the weight of each data in each divided park using the historical regression analysis method, each data is multiplied by the corresponding weight using a multiplication operation and the sum is calculated to obtain the comprehensive service demand intensity of each divided park.

[0021] Sort the comprehensive service demand intensity in ascending order, use the empirical rule to set the number of selections, traverse the sorting results, and gradually select the largest comprehensive service demand intensity based on the number of selections. Define the divided park corresponding to the comprehensive service demand intensity as the candidate facility location, traverse each candidate facility location, and randomly select a service point. Use the Euclidean distance formula to calculate the distance from the service point to all candidate facility locations.

[0022] Use the Voronoi diagram algorithm combined with distance to segment each candidate facility location and obtain the Voronoi diagram service area of ​​each facility. In the Voronoi diagram service area, if the current area and the adjacent area have a shared boundary, the area pair is marked as directly reachable, otherwise it is marked as infinite.

[0023] Use the Euclidean distance formula to calculate the distance between pairs of facilities in the directly reachable area, and use it to construct a distance matrix in combination with infinity. Use the NetworkX tool to construct a directed graph in combination with the distance matrix.

[0024] Use the Floyd algorithm to extract the outgoing and incoming edges of each facility from the directed graph, and obtain the outgoing and incoming edge sets of each facility;

[0025] According to the outgoing edge set and the incoming edge set, calculate the product of the number of outgoing edges and incoming edges in the set;

[0026] After the products are sorted in ascending order, the facilities corresponding to the products are selected as intermediate nodes according to the sorting results for iteration. During the iteration process, the distance between the facility pairs is defined as the weight. When the facility pairs can be reached through the intermediate nodes, the path is updated through the relaxation operation.

[0027] When the distance between the facility pair (e, s) in the updated path is updated from infinity to a finite value, it means that the path of the facility pair (e, s) is updated, and the e-th facility is added to the incoming edge set of the intermediate node k, and the s-th facility is added to the outgoing edge set of the intermediate node k;

[0028] Continue iterative updating, and output the shortest path matrix after the number of iterations reaches the maximum number;

[0029] The shortest path matrix refers to the shortest physical distance between each pair of facilities in the park.

[0030] As a preferred solution of the intelligent park integrated management system based on intelligent data collection described in the present invention, the calculation of facility priority scores, combining the shortest distance matrix to generate the facility start-stop sequence and scheduling individuals for iterative optimization to obtain an initial solution includes the following steps:

[0031] The weight coefficient is set using the historical regression analysis method, and the priority score of each facility is calculated based on the facility energy consumption and the minimum distance value of the facility in the shortest path matrix. Based on the priority score, all facilities are sorted in descending order according to the priority score to obtain the facility arrangement result, which is defined as the start-stop order;

[0032] Use the domain method to set the adjustment coefficient value range [ψ min ,ψ max ], where ψ minrepresents the minimum adjustment coefficient, ψ max Indicates the maximum adjustment coefficient;

[0033] Use a random number generator to generate adjustment coefficients within a range of values ​​and calculate the power of each facility;

[0034] Use coding technology to encode the start and stop sequence and power and then combine them to generate scheduling individuals. Randomly generate a population for initialization, define the objective function, calculate the objective function value, use the objective function value as the individual fitness value and sort them in ascending order. Select the largest fitness value for crossover and mutation operations, generate candidate solutions and iterate. When the number of iterations reaches the maximum number, stop the iteration and output the initial solution.

[0035] The initial solution refers to the start and stop sequence and power setting combination of lighting and ventilation facilities.

[0036] As a preferred solution of the intelligent park integrated management system based on intelligent collection described in the present invention, wherein: generating elite and common populations based on the initial solution for initialization, and optimizing through perturbation, path selection probability and pheromone diffusion mechanism, combined with Hamming distance difference evaluation and cross-replacement strategy, to obtain the optimal solution includes the following steps:

[0037] The initial solution is used as the elite ant individual, and Gaussian perturbation and mutation operations are used to perturb and mutate the elite ant individual to generate an elite population for initialization. θ elite ant individuals are randomly selected from the elite population and defined as ordinary ant individuals. Combined with random generation operations, an ordinary population is generated for initialization. The inverse of the initial solution objective function value is defined as heuristic information, and the Jess rule is used to set pheromones.

[0038] The cross-validation method is used to set the weights of heuristic information and pheromone respectively, and the statistical method is used to count all the adjacent facilities of the current facility and construct the adjacent facility set;

[0039] Use the path selection probability formula to calculate the path selection probability;

[0040] Use roulette to select the path selection probability, and take the path corresponding to the selected path selection probability as the current path;

[0041] Use the analytic hierarchy process to set the weight coefficients of lighting and ventilation facilities and define the objective function;

[0042] Calculate the objective function value and use it as the fitness value of the elite or ordinary ant individual; perform iterative updates. During the iteration process, use the Hamming distance to calculate the starting state distance between the elite ant individual and the adjacent ordinary ant individual, which is defined as the starting state difference;

[0043] The Hamming distance is further used to calculate the endpoint state distance between the elite ant individual and the adjacent ordinary ant individual, which is defined as the endpoint state difference. The addition formula is used to calculate the sum of the difference between the starting point and the end point, which is defined as the comprehensive difference. The constant is set using the empirical rule, and the pheromone diffusion is updated based on the comprehensive difference.

[0044] The field method is further used to set the exchange iteration number and the selection number. When the number of iterations reaches the exchange iteration number, the elite and ordinary populations are sorted in ascending order according to the fitness value, and the first i ordinary individuals in the ordinary population arrangement result are selected from the selection number for cross-replacement with the first i elite individuals in the elite population arrangement result. After the cross-replacement, the grid search method is used to set the attenuation factor, and the state path is fused. The fused state path is used as the new individual, and the global pheromone update is performed.

[0045] Continue iterating and updating. When the number of iterations reaches the maximum, stop iterating and output the optimal solution.

[0046] The optimal solution includes the start and stop sequence and power setting combination of the lighting and ventilation equipment in all time slices.

[0047] As a preferred solution of the smart park integrated management system based on intelligent data collection described in the present invention, the construction of an evaluation model to predict facility status and triggering optimal solution recalculation and early warning based on the prediction results include the following steps:

[0048] Collect historical operating data of lighting and ventilation facilities;

[0049] The historical operation data includes current, temperature and working status;

[0050] An evaluation model was constructed using a multi-layer perceptron architecture, with mean square error defined as the objective function. Grid search and cross-validation were used to optimize model parameters, taking the optimized parameters and historical operating data as inputs to output the predicted status value for each facility.

[0051] Use IF-THEN rules to set the warning threshold, compare the state prediction value with the warning threshold, and when the current state is greater than or equal to the warning threshold, the facility will be eliminated and the optimal solution will be recalculated and the warning will be triggered. Otherwise, monitoring will continue.

[0052] As a preferred solution of the smart park comprehensive management system based on intelligent collection described in the present invention, the implementation of corresponding measures refers to generating a report on the comparison results and optimization suggestions when triggering an early warning and continuing monitoring, sending them to the management personnel through a transmission protocol, and implementing corresponding measures through the management personnel.

[0053] As a preferred solution of the smart park comprehensive management system based on intelligent collection described in the present invention, after collecting the facility status, environment and traffic data, preprocessing operations are first performed, including denoising, interpolation and normalization.

[0054] As a preferred solution for the smart park comprehensive management system based on intelligent collection described in the present invention, the display of data through a visual interface refers to using the Power tool to convert the comparison results into charts, and visually displaying the comparison results, optimization suggestions and optimal solutions through charts.

[0055] As a preferred solution of the smart park comprehensive management system based on intelligent collection described in the present invention, the use of a database for data storage refers to storing the comparison results through a MySQL database, and adding facility ID and timestamp to the comparison results through the MySQL database for classification.

[0056] As a preferred solution of the smart park comprehensive management system based on intelligent collection described in the present invention, the encryption by encryption algorithm refers to encrypting the stored data using a symmetric encryption algorithm and regularly changing the key.

[0057] The beneficial effects of the present invention are as follows: by introducing the Voronoi diagram algorithm and the Floyd algorithm and calculating the shortest physical distance between each pair of facilities, the present invention can accurately adjust the location of the facilities, make the facility layout of the park more reasonable, maximize the service scope, thereby improving the overall service quality and operating efficiency; secondly, through the intelligent scheduling of the ant colony algorithm, the present invention can automatically optimize the start-stop sequence and power setting of the facilities in the park according to the energy consumption, service demand and distance relationship of each facility in the park, thereby improving the operating efficiency of the facilities in the park, reducing energy waste, and ensuring the reasonable distribution of facility loads. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] 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.

[0059] Figure 1 This is a structural diagram of the smart park comprehensive management system based on intelligent collection in Example 1.

[0060] Figure 2 This is a flowchart of the calculation and iteration module in Example 1.

[0061] Figure 3 This is a flow chart of the optimization module in Example 1. DETAILED DESCRIPTION

[0062] 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.

[0063] 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.

[0064] 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.

[0065] Example 1, reference Figures 1 to 3 , which is the first embodiment of the present invention, provides a smart park integrated management system based on intelligent collection, including the following steps:

[0066] S1, the collection and generation module collects raw data, segments the service area using the Voronoi diagram, and combines it with the Floyd algorithm to generate a shortest distance matrix; the raw data includes the facility status of the lighting and ventilation facilities in the park and the park environment data;

[0067] Specifically, collecting raw data, segmenting the service area using the Voronoi diagram, and combining the Floyd algorithm to generate the shortest distance matrix include the following steps:

[0068] Use IoT sensors to collect real-time data on the status of lighting and ventilation facilities within the park and on the park environment;

[0069] The facility status data includes the switch status, energy consumption and location of lighting and ventilation facilities;

[0070] The environmental data includes temperature and light intensity;

[0071] The park is divided into sections using the grid division method. After setting the weight of each data in each divided park using the historical regression analysis method, each data is multiplied by the corresponding weight using a multiplication operation and the sum is calculated to obtain the comprehensive service demand intensity of each divided park.

[0072] The comprehensive service demand intensity determines the operational demand for facilities in the area. For example, if the first area scores 10 and the second area scores 15, the second area has a higher operational demand for facilities.

[0073] Sort the comprehensive service demand intensity in ascending order, use the empirical rule to set the number of selections, traverse the sorting results, and gradually select the largest comprehensive service demand intensity based on the number of selections. Define the zone corresponding to the comprehensive service demand intensity as the candidate facility location;

[0074] Iterate over each candidate facility location and randomly select a service point. Use the Euclidean distance formula to calculate the distance from the service point to all candidate facility locations.

[0075] Use the Voronoi diagram algorithm combined with distance to segment each candidate facility location and obtain the Voronoi diagram service area of ​​each facility;

[0076] In the Voronoi diagram service area, if the current area and the adjacent area have a shared boundary, the area pair is marked as directly reachable, otherwise it is marked as infinite;

[0077] Use the Euclidean distance formula to calculate the distance between pairs of facilities in the directly reachable area, and use it to construct a distance matrix in combination with infinity. Use the NetworkX tool to construct a directed graph in combination with the distance matrix.

[0078] Use the Floyd algorithm to extract the outgoing and incoming edges of each facility from the directed graph, and obtain the outgoing and incoming edge sets of each facility;

[0079] The difference between outgoing and incoming edges lies in the direction of the path in a directed graph;

[0080] Outgoing edges refer to the paths from the current facility to other facilities;

[0081] Incoming edges refer to the paths between other facilities and the current facility;

[0082] According to the outgoing edge set and the incoming edge set, the product of the number of outgoing edges and incoming edges in the set is calculated. The formula is:

[0083] A1(e)=|in[e]|×|out[e]|

[0084] Where A1(e) represents the product of the number of outgoing and incoming edges of the e-th facility, |·| represents the absolute value, in[e] represents the incoming edge set of the e-th facility, and out[e] represents the outgoing edge set of the e-th facility.

[0085] After arranging the products in ascending order, the facilities corresponding to the products are selected as intermediate nodes for iteration according to the arrangement results;

[0086] In the iterative process, the distance between the facility pairs is defined as the weight. When the facility pairs can be reached through the intermediate nodes, the path is updated through the relaxation operation. The formula is:

[0087]

[0088] Where, represents the shortest distance between the e-th facility and the s-th facility after transiting through the intermediate node k, represents the distance between the e-th facility and the s-th facility at the m-1th iteration, represents the distance between the e-th facility and the intermediate node k at the m-1th iteration, represents the distance between the intermediate node k and the sth facility at the m-1th iteration, w es represents the weight of the e-th facility and the s-th facility, min(·) represents the minimum value operation, represents the incoming edge set of the intermediate node k in the m-1th iteration, represents the outgoing edge set of the intermediate node k in the m-1th iteration, ∈ represents belonging;

[0089] When the distance between the facility pair (e, s) in the updated path is updated from infinity to a finite value, it means that the path of the facility pair (e, s) is updated, and the e-th facility is added to the incoming edge set of the intermediate node k, and the s-th facility is added to the outgoing edge set of the intermediate node k;

[0090] Continue iterative updating, and output the shortest path matrix after the number of iterations reaches the maximum number;

[0091] The shortest path matrix refers to the shortest physical distance between each pair of facilities in the park.

[0092] IoT sensors collect status data and environmental data of lighting and ventilation facilities in the park in real time, so that park managers can grasp the energy consumption, environmental conditions and facility operation status of the park in real time. The grid division method is combined with the historical regression analysis method to calculate the comprehensive service demand intensity, providing a scientific basis for facility layout and resource scheduling. The service demand intensity assessment of each park area makes the distribution of park facilities more reasonable and ensures the optimal utilization of resources. The Voronoi diagram is used to segment spatial areas, and the Floyd algorithm is used to optimize the paths between facilities. Combining the advantages of the two, it is possible to optimize the distance between facilities and the division of service areas on the basis of ensuring coverage of park needs, thereby improving the facilities in the park. The response speed and energy efficiency effectively avoid the traditional single layout method, and enable the present invention to dynamically respond to environmental changes and facility needs. By optimizing the distance matrix through the Floyd algorithm, the shortest path between facilities in the park can be accurately calculated, which enables the present invention to reduce redundant operations between facilities and ensure that each facility provides services at the most appropriate time, thereby minimizing energy consumption and improving the overall efficiency of park facilities. Secondly, the relaxation operation ensures that the shortest path calculated each time is the most optimized. This process not only improves the accuracy of path calculation, but also ensures that the interconnection between facilities is more efficient by continuously correcting erroneous paths, so that the present invention can avoid delays or energy waste caused by improper path selection.

[0093] S2, a calculation and iteration module, calculates the facility priority scores, generates the facility start-stop sequence and schedules the individual facilities for iterative optimization based on the shortest distance matrix, and obtains an initial solution; the initial solution includes the start-stop sequence and power setting combination;

[0094] Specifically, the facility priority scores are calculated, and the start and stop sequence of the facilities is generated by combining the shortest distance matrix and the individual dispatchers are iteratively optimized to obtain the initial solution, which includes the following steps:

[0095] The weight coefficient is set using the historical regression analysis method, and the priority score of each facility is calculated based on the facility energy consumption and the minimum distance value of the facility in the shortest path matrix. The formula is:

[0096]

[0097] Where S e represents the priority score of the e-th facility, Represents the weight coefficient of energy consumption of the e-th facility, E e represents the energy consumption of the e-th facility, The weight coefficient representing the minimum distance value, d es represents the shortest distance between the e-th facility and the s-th facility;

[0098] According to the priority scores, all facilities are sorted in descending order according to the priority scores to obtain the facility arrangement result, which is defined as the start-stop order;

[0099] Use the domain method to set the adjustment coefficient value range [ψ min ,ψ max ], where ψ min represents the minimum adjustment coefficient, ψ max Indicates the maximum adjustment coefficient;

[0100] A random number generator is used to generate an adjustment coefficient within a range of values ​​and the power of each facility is calculated using the formula:

[0101]

[0102] Where G e represents the power of the e-th facility, θ i represents the adjustment coefficient of the e-th facility, represents the maximum power of the e-th facility, which can be obtained from the facility manual;

[0103] Use coding technology to encode the start and stop sequence and power and then combine them to generate scheduling individuals, and randomly generate a population for initialization;

[0104] Define the objective function, the formula is:

[0105]

[0106] Where C represents the objective function value, n represents the total number of facilities, and t e Indicates the start and stop sequence of the e-th facility;

[0107] Calculate the objective function value, use it as the individual fitness value and sort them in ascending order, select the largest fitness value for crossover and mutation operations, generate candidate solutions for iteration, and stop iteration when the maximum number of iterations is reached, and output the initial solution;

[0108] The initial solution refers to the start and stop sequence and power setting combination of lighting and ventilation facilities.

[0109] By combining the weight coefficients set by the historical regression analysis method with the shortest distance matrix, the priority score of each facility can be accurately calculated, which ensures that the scheduling order of the facilities not only takes into account energy consumption, but also takes into account factors such as the geographical location and functional requirements between facilities, thereby achieving an optimized start-stop sequence. This step enables the present invention to effectively avoid the problem of ignoring the interaction and distance factors between facilities in traditional scheduling schemes, making the scheduling process more scientific and reasonable. The facilities are sorted according to the priority scores, the start-stop sequence is generated, and the power of the facilities is flexibly controlled by adjusting the coefficients. This enables the facilities to operate at optimal power in different operating time periods and environmental conditions, thereby achieving the goal of energy conservation and emission reduction. By introducing a dynamic adjustment mechanism, the power setting of the facility can be adjusted at any time according to the actual needs of the facility, avoiding the energy waste that may be caused by fixed power settings. Or insufficient service, and the start and stop sequence and power settings are encoded using coding technology, and through population initialization and iterative optimization as well as multiple crossover and mutation operations, not only the adaptability of the present invention is improved, but also the globality of the optimization is enhanced. The iterative optimization mechanism can adaptively adjust the scheduling plan of the facility to provide the best solution under different operating conditions. Secondly, by optimizing the start and stop sequence and power settings of the facility, the efficiency of energy management can be significantly improved. Through reasonable scheduling, unnecessary operating time and power consumption of the equipment are reduced, and energy waste is reduced. In addition, the present invention can ensure that the facility operates in the optimal state by optimizing the power setting in real time, thereby improving user comfort. For example, lighting and ventilation facilities can automatically adjust their operating modes according to changes in environmental conditions, providing a more comfortable and energy-saving use environment.

[0110] S3, the optimization module, generates elite and common populations based on the initial solution for initialization, and optimizes through perturbation, path selection probability, and pheromone diffusion mechanism, combined with Hamming distance difference evaluation and cross-replacement strategy to obtain the optimal solution; the optimal solution includes the optimized start-stop sequence and power setting combination;

[0111] Specifically, based on the initial solution, elite and ordinary populations are generated for initialization, and optimization is performed through perturbation, path selection probability, and pheromone diffusion mechanism, combined with Hamming distance difference evaluation and cross-replacement strategy. The optimal solution includes the following steps:

[0112] The initial solution is used as the elite ant individual, and Gaussian perturbation and mutation operations are used to perturb and mutate the elite ant individual to generate an elite population for initialization;

[0113] Randomly select θ elite ant individuals from the elite population, define them as ordinary ant individuals, and combine them with random generation operations to generate an ordinary population for initialization;

[0114] The inverse of the initial solution objective function value is defined as heuristic information, and the pheromone is set using the Jess rule;

[0115] The cross-validation method was used to set the weights of heuristic information and pheromone respectively;

[0116] Use statistical methods to count all adjacent facilities of the current facility and construct an adjacent facility set;

[0117] Use the path selection probability formula to calculate the path selection probability. The formula is:

[0118]

[0119] Where H es represents the path selection probability from the e-th facility to the s-th facility, represents the pheromone from the e-th facility to the s-th facility, η es represents the heuristic information from the e-th facility to the s-th facility, represents the weight of pheromone, ξ represents the weight of heuristic information, N e represents the set of adjacent facilities of the e-th facility, represents the pheromone from the eth facility to the oth facility, η eo Represents the heuristic information from the e-th facility to the o-th facility;

[0120] Use roulette to select the path selection probability, and take the path corresponding to the selected path selection probability as the current path;

[0121] Use the analytic hierarchy process to set the weight coefficients of lighting and ventilation facilities;

[0122] Define the objective function and minimize the objective function value. The formula is:

[0123]

[0124] Where F(r) represents the objective function value at time r, T represents the total time, α represents the weight coefficient of lighting facilities, y represents the total number of lighting facilities, G e (r) represents the energy consumption of the e-th lighting facility at time r, represents the weight coefficient of ventilation facilities, g represents the total number of ventilation facilities, J s (r) represents the energy consumption of the sth facility in the ventilation facilities at time r;

[0125] Calculate the objective function value and use it as the fitness value of the elite or ordinary ant individual; perform iterative updates. During the iteration process, use the Hamming distance to calculate the starting state distance between the elite ant individual and the adjacent ordinary ant individual, which is defined as the starting state difference. The formula is:

[0126]

[0127] Where, f1(S a ,S b ) represents the difference in starting state S between the ath elite ant individual and the bth elite ant individual, n represents the total number of facilities, represents the start and stop state of the u-th facility in the starting state S of the a-th elite ant individual, represents the start and stop status of the u-th facility in the starting state S of the b-th elite ant individual;

[0128] The Hamming distance is further used to calculate the terminal state distance between the elite ant individual and the adjacent ordinary ant individual, which is defined as the terminal state difference. The addition formula is used to calculate the sum of the difference between the starting point and the terminal point, which is defined as the comprehensive difference.

[0129] Use the empirical rule to set the constant and combine it with the comprehensive difference to update the diffusion of pheromones. The formula is:

[0130]

[0131] Where, represents the diffusion update pheromone of the ath elite ant individual and the bth elite ant individual in the t+1th round of iteration, Q represents a constant, represents the pheromone of the ath elite ant individual and the bth elite ant individual in the tth iteration. f3(a,b)<1 means that the pheromone diffusion update is performed when the comprehensive difference between the ath elite ant individual and the bth elite ant individual is less than 1. f3(a,b)≥1 means that the pheromone diffusion update is not performed when the comprehensive difference between the ath elite ant individual and the bth elite ant individual is greater than or equal to 1.

[0132] The domain method is further used to set the exchange iteration number and the selection number. When the number of iterations reaches the exchange iteration number, the elite and ordinary populations are sorted in ascending order according to the fitness value, and the first i ordinary individuals in the ordinary population are selected from the arrangement result of the ordinary population according to the selection number to cross-replace with the first i elite individuals in the arrangement result of the elite population.

[0133] After cross-replacement, the grid search method is used to set the attenuation factor and perform state path fusion. The formula is:

[0134]

[0135] Where S y (t) represents the yth new path integrated in the tth iteration, ω represents the weight factor, represents the zth optimal state path at the tth iteration, represents the vth elite path in the tth iteration;

[0136] The fused state path is used as a new individual and the global pheromone update is performed. The formula is:

[0137]

[0138] Where Δv R (t) represents the global pheromone increment at the tth iteration, S R (t) represents the increment of the global optimal path at the tth iteration, F3(t) represents the objective function value of the global optimal path at the tth iteration, S I (t) represents the pheromone increment of the optimal path in the tth round of iteration, F2(t) represents the objective function value of the optimal path in the tth round of iteration, represents the global pheromone updated by the a-th elite ant individual and the b-th elite ant individual in the t-th iteration, ρ R represents the attenuation factor, represents the initial global pheromone of the a-th elite ant individual and the b-th elite ant individual in the t-th iteration;

[0139] Continue iterating and updating. When the number of iterations reaches the maximum, stop iterating and output the optimal solution.

[0140] The optimal solution includes the start and stop sequence and power setting combination of the lighting and ventilation equipment in all time slices.

[0141] For example, lighting facilities:

[0142] Facility 2: Open during the day (8:00-18:00), power 80%, closed at night;

[0143] Facility 1: Open in the evening (17:00-22:00), power 60%;

[0144] Facility 3: On all day, 50% power;

[0145] Ventilation facilities:

[0146] Facility 1: Open during the high temperature period (12:00-16:00), power 70%;

[0147] Facility 2: On all day, 50% power.

[0148] Through the dynamic adjustment of elite and ordinary populations, combined with the Hamming distance evaluation difference, the quality of the solution can be continuously improved in multiple iterations, so that the final solution is closer to the global optimum. The elite population generated by the elite ant individuals after perturbation and mutation operations can find a better path in the solution space, while the ordinary population generated by ordinary ant individuals through random operations increases the diversity of solutions and helps to avoid falling into local optimal solutions. By accurately calculating the energy consumption of lighting and ventilation facilities and combining them with environmental data, the present invention can achieve dual optimization of lighting and ventilation facilities. The weight coefficients of different facilities are set by the hierarchical analysis method, which ensures the reasonable allocation of lighting and ventilation facilities under different operating conditions, thereby maximizing the efficiency. Energy efficiency is improved, and through pheromone diffusion and path selection probability mechanisms, the algorithm can make real-time adjustments based on current environmental changes and facility status, rather than relying on fixed rules. This real-time adjustment capability gives the present invention significant advantages in dealing with actual complex dynamic systems, and can optimize the start and stop status and power settings of facilities at different times and under different environmental conditions. By accurately optimizing the energy consumption of lighting and ventilation facilities, it can not only reduce unnecessary energy consumption but also reduce carbon emissions, which is of great significance to environmental protection. For example, in modern urban and industrial environments, improving energy efficiency is crucial to achieving sustainable development goals. The present invention can provide intelligent decision-making support for facility management and reduce the complexity of manual intervention and adjustment.

[0149] S4, early warning module, builds an evaluation model to predict the status of facilities, triggers optimal solution recalculation and early warning according to the prediction results, and implements corresponding measures;

[0150] Specifically, building an evaluation model to predict facility status and triggering optimal solution recalculation and early warning based on the prediction results includes the following steps:

[0151] Collect historical operating data of lighting and ventilation facilities;

[0152] The historical operation data includes current, temperature and working status;

[0153] The evaluation model was constructed using a multi-layer perceptron architecture, with mean square error defined as the objective function. Grid search and cross-validation were used to optimize the model parameters.

[0154] Taking the optimization parameters and historical operating data as input, the output is the status prediction value of each facility;

[0155] Use IF-THEN rules to set the warning threshold, compare the state prediction value with the warning threshold, and when the current state is greater than or equal to the warning threshold, the facility will be eliminated and the optimal solution will be recalculated and the warning will be triggered. Otherwise, monitoring will continue.

[0156] Through deep learning analysis of historical data, the future status of the facility can be accurately predicted, providing data support for decision-making, so that the present invention can not only reduce facility failures, but also optimize the operation mode of the facility, thereby improving energy efficiency. By predicting equipment failures or abnormal conditions in advance, timely warnings can be issued to avoid downtime and high maintenance costs caused by failures. Secondly, after setting a reasonable warning threshold, the present invention can monitor the facility status in real time and issue a warning before the abnormality occurs. The warning mechanism can take maintenance or adjustment measures in advance to avoid human omissions and response delays. In addition, through real-time status prediction and monitoring, the present invention can promptly discover potential safety hazards in equipment operation and reduce the occurrence of sudden accidents.

[0157] Furthermore, implementing corresponding measures means that when an early warning is triggered and monitoring is continued, a report of the comparison results and optimization suggestions are generated, sent to management personnel through a transmission protocol, and corresponding measures are implemented by the management personnel.

[0158] After generating a report on the comparison results and optimization suggestions, the report is sent to the management personnel through the transmission protocol, and the management personnel implement corresponding measures. This allows the present invention to provide the management personnel with a clear direction for improvement. With the help of an efficient transmission protocol, the information is quickly transmitted to the management personnel, allowing the management personnel to respond quickly based on the data.

[0159] S5, data management module, displays data through a visual interface, and uses a database to store data and encrypts it using an encryption algorithm;

[0160] Specifically, displaying data through a visual interface means using the Power tool to convert the comparison results into charts, and visually displaying the comparison results, optimization suggestions, and optimal solutions through the charts.

[0161] By converting the comparison results into charts, the data display becomes more intuitive, and the comparison results, optimization suggestions and optimal solutions are displayed visually, allowing managers to quickly capture the differences, trends and potential problems between the data.

[0162] Furthermore, using a database for data storage means storing the comparison results through a MySQL database, and adding a facility ID and a timestamp to the comparison results through the MySQL database before classification.

[0163] By using a MySQL database for data storage and adding facility IDs and timestamps to comparison results, efficient storage, accurate classification, and flexible historical data management are achieved.

[0164] Furthermore, encryption through an encryption algorithm refers to encrypting stored data using a symmetric encryption algorithm and periodically changing the key.

[0165] By combining symmetric encryption algorithms with regular key replacement, the efficiency and security of data encryption are improved. Moreover, through the key replacement mechanism, adaptive protection is provided in dynamic environments, enhancing the reliability and flexibility of the encryption system.

[0166] 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 smart park integrated management system based on intelligent data collection, characterized by: include, The acquisition and generation module collects raw data, segments the service area using the Voronoi diagram, and combines the Floyd algorithm to generate the shortest distance matrix; The original data includes the facility status of the lighting and ventilation facilities in the park and the park environment data; The calculation and iteration module calculates the facility priority scores, generates the facility start and stop sequence based on the shortest distance matrix, and performs iterative optimization on the scheduling individuals to obtain the initial solution; The initial solution includes a start-stop sequence and a power setting combination; The optimization module generates elite and common populations based on the initial solution for initialization, and optimizes through perturbation, path selection probability, and pheromone diffusion mechanism, combined with Hamming distance difference evaluation and cross-replacement strategy to obtain the optimal solution; The optimal solution includes an optimized start-stop sequence and power setting combination; The early warning module builds an evaluation model to predict the status of facilities, triggers optimal solution recalculation and early warning based on the prediction results, and implements corresponding measures; The data management module displays the data through a visual interface, uses the database to store the data, and then encrypts it using an encryption algorithm.

2. The intelligent park integrated management system based on intelligent collection as claimed in claim 1, characterized in that: The process of collecting raw data, segmenting the service area using a Voronoi diagram, and generating a shortest distance matrix using the Floyd algorithm includes the following steps: Use IoT sensors to collect real-time data on the status of lighting and ventilation facilities within the park and on the park environment; The facility status data includes the switch status, energy consumption and location of lighting and ventilation facilities; The environmental data includes temperature and light intensity; The park is divided into sections using the grid division method. After setting the weight of each data in each divided park using the historical regression analysis method, each data is multiplied by the corresponding weight using a multiplication operation and the sum is calculated to obtain the comprehensive service demand intensity of each divided park. Sort the comprehensive service demand intensity in ascending order, use the empirical rule to set the number of selections, traverse the sorting results, and gradually select the largest comprehensive service demand intensity based on the number of selections. Define the divided park corresponding to the comprehensive service demand intensity as the candidate facility location, traverse each candidate facility location, and randomly select a service point. Use the Euclidean distance formula to calculate the distance from the service point to all candidate facility locations. Use the Voronoi diagram algorithm combined with distance to segment each candidate facility location and obtain the Voronoi diagram service area of ​​each facility. In the Voronoi diagram service area, if the current area and the adjacent area have a shared boundary, the area pair is marked as directly reachable, otherwise it is marked as infinite. Use the Euclidean distance formula to calculate the distance between pairs of facilities in the directly reachable area, and use it to construct a distance matrix in combination with infinity. Use the NetworkX tool to construct a directed graph in combination with the distance matrix. Use the Floyd algorithm to extract the outgoing and incoming edges of each facility from the directed graph, and obtain the outgoing and incoming edge sets of each facility; According to the outgoing edge set and the incoming edge set, calculate the product of the number of outgoing edges and incoming edges in the set; After the products are sorted in ascending order, the facilities corresponding to the products are selected as intermediate nodes according to the sorting results for iteration. During the iteration process, the distance between the facility pairs is defined as the weight. When the facility pairs can be reached through the intermediate nodes, the path is updated through the relaxation operation. When the distance between the facility pair (e, s) in the updated path is updated from infinity to a finite value, it means that the path of the facility pair (e, s) is updated, and the e-th facility is added to the incoming edge set of the intermediate node k, and the s-th facility is added to the outgoing edge set of the intermediate node k; Continue iterative updating, and output the shortest path matrix after the number of iterations reaches the maximum number; The shortest path matrix refers to the shortest physical distance between each pair of facilities in the park.

3. The intelligent park integrated management system based on intelligent collection as claimed in claim 2, characterized in that: The calculation of facility priority scores, combined with the shortest distance matrix to generate the facility start and stop sequence and the scheduling individuals for iterative optimization to obtain an initial solution includes the following steps: The weight coefficient is set using the historical regression analysis method, and the priority score of each facility is calculated based on the facility energy consumption and the minimum distance value of the facility in the shortest path matrix. Based on the priority score, all facilities are sorted in descending order according to the priority score to obtain the facility arrangement result, which is defined as the start-stop order; Use the domain method to set the adjustment coefficient value range [ψ min ,ψ max ], where ψ min represents the minimum adjustment coefficient, ψ max Indicates the maximum adjustment coefficient; Use a random number generator to generate adjustment coefficients within a range of values ​​and calculate the power of each facility; Use coding technology to encode the start and stop sequence and power and then combine them to generate scheduling individuals. Randomly generate a population for initialization, define the objective function, calculate the objective function value, use the objective function value as the individual fitness value and sort them in ascending order. Select the largest fitness value for crossover and mutation operations, generate candidate solutions and iterate. When the number of iterations reaches the maximum number, stop the iteration and output the initial solution. The initial solution refers to the start and stop sequence and power setting combination of lighting and ventilation facilities.

4. The intelligent park integrated management system based on intelligent collection as claimed in claim 3, characterized in that: The method of generating elite and common populations based on the initial solution for initialization, and optimizing through perturbation, path selection probability, and pheromone diffusion mechanism, combined with Hamming distance difference evaluation and cross-replacement strategy, to obtain the optimal solution includes the following steps: The initial solution is used as the elite ant individual, and Gaussian perturbation and mutation operations are used to perturb and mutate the elite ant individual to generate an elite population for initialization. θ elite ant individuals are randomly selected from the elite population and defined as ordinary ant individuals. Combined with random generation operations, an ordinary population is generated for initialization. The inverse of the initial solution objective function value is defined as heuristic information, and the Jess rule is used to set pheromones. The cross-validation method is used to set the weights of heuristic information and pheromone respectively, and the statistical method is used to count all the adjacent facilities of the current facility and construct the adjacent facility set; Use the path selection probability formula to calculate the path selection probability; Use roulette to select the path selection probability, and take the path corresponding to the selected path selection probability as the current path; Use the analytic hierarchy process to set the weight coefficients of lighting and ventilation facilities and define the objective function; Calculate the objective function value and use it as the fitness value of the elite or ordinary ant individual; perform iterative updates. During the iteration process, use the Hamming distance to calculate the starting state distance between the elite ant individual and the adjacent ordinary ant individual, which is defined as the starting state difference; The Hamming distance is further used to calculate the endpoint state distance between the elite ant individual and the adjacent ordinary ant individual, which is defined as the endpoint state difference. The addition formula is used to calculate the sum of the difference between the starting point and the end point, which is defined as the comprehensive difference. The constant is set using the empirical rule, and the pheromone diffusion is updated based on the comprehensive difference. The field method is further used to set the exchange iteration number and the selection number. When the number of iterations reaches the exchange iteration number, the elite and ordinary populations are sorted in ascending order according to the fitness value, and the first i ordinary individuals in the ordinary population arrangement result are selected from the selection number for cross-replacement with the first i elite individuals in the elite population arrangement result. After the cross-replacement, the grid search method is used to set the attenuation factor, and the state path is fused. The fused state path is used as the new individual, and the global pheromone update is performed. Continue iterating and updating. When the number of iterations reaches the maximum, stop iterating and output the optimal solution. The optimal solution includes the start and stop sequence and power setting combination of the lighting and ventilation equipment in all time slices.

5. The intelligent park integrated management system based on intelligent collection as claimed in claim 4, characterized in that: The construction of the evaluation model to predict the facility status and trigger the optimal solution recalculation and early warning according to the prediction results include the following steps: Collect historical operating data of lighting and ventilation facilities; The historical operation data includes current, temperature and working status; An evaluation model was constructed using a multi-layer perceptron architecture, with mean square error defined as the objective function. Grid search and cross-validation were used to optimize model parameters, taking the optimized parameters and historical operating data as inputs to output the predicted status value for each facility. Use IF-THEN rules to set the warning threshold, compare the state prediction value with the warning threshold, and when the current state is greater than or equal to the warning threshold, the facility will be eliminated and the optimal solution will be recalculated and the warning will be triggered. Otherwise, monitoring will continue.

6. The smart park integrated management system based on intelligent data collection as claimed in claim 5, characterized in that: The implementation of corresponding measures refers to generating a report of the comparison results and optimization suggestions when triggering an early warning and continuing monitoring, sending them to management personnel through a transmission protocol, and having the management personnel implement corresponding measures.

7. The intelligent park integrated management system based on intelligent collection as claimed in claim 1, characterized in that: After collecting the facility status, environment and traffic data, preprocessing operations are first performed, including denoising, interpolation and normalization.

8. The intelligent park integrated management system based on intelligent collection as claimed in claim 6, characterized in that: The display of data through a visual interface refers to using the Power tool to convert the comparison results into charts, and visually displaying the comparison results, optimization suggestions and optimal solutions through the charts.

9. The smart park integrated management system based on intelligent data collection as claimed in claim 8, characterized in that: The use of a database for data storage refers to storing the comparison results through a MySQL database, and adding a facility ID and a timestamp to the comparison results through the MySQL database before classification.

10. The smart park integrated management system based on intelligent collection as claimed in claim 9, characterized in that: The encryption by an encryption algorithm refers to encrypting the stored data using a symmetric encryption algorithm and regularly changing the key.

Images