Intelligent park equipment maintenance optimization system based on digital twin model

Abstract

The application discloses a smart park equipment maintenance optimization system based on a digital twin model, relates to the technical field of intelligent optimization, and comprises a division module, a construction module and a regulation and control module, divides equipment groups, calculates a first correlation degree, obtains a first power path through distribution, optimizes a second power path, and executes a first operation according to a first judgment result. According to the application, equipment groups are divided according to equipment purposes, so that equipment management is more targeted. Through calculation of the first correlation degree, management efficiency is improved, the improved genetic algorithm is used to optimize the power path, the second power path is obtained, more accurate and efficient energy distribution is realized, energy consumption is reduced, the system helps the park to realize energy saving and emission reduction, further reduces operation cost, integrates scattered equipment and data, breaks the information island between systems in the traditional park, realizes intelligent linkage between systems, and improves the digital management capability of the park.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of intelligent optimization, and in particular to a smart park equipment maintenance optimization system based on a digital twin model. BACKGROUND

[0002] In recent years, a digital twin model can deeply fuse various systems in a park, such as a security system, an energy management system and a device monitoring system, break information islands, realize data sharing and collaborative work, utilize three-dimensional modeling technology, and present device layout and building structure information of the park in an intuitive three-dimensional visual manner. Through widely deploying a sensor network in the park, the digital twin model can collect multi-dimensional information such as device operation data and environment data in real time, and map the multi-dimensional information to a virtual model to realize real-time monitoring and dynamic perception of the device operation state.

[0003] At present, a kind of after-sales equipment predictive maintenance collaborative system based on digital twinning is disclosed in Chinese patent application CN112418523A. The method collects and transmits the operation state parameters of the equipment through the collaborative maintenance seven-dimensional model of digital twinning. The collaborative maintenance seven-dimensional model performs predictive analysis on the collected equipment operation parameters to obtain the prediction result of the equipment. According to the obtained equipment prediction result, it is judged whether the current equipment is normal. When it is judged that the current equipment is in an abnormal state, the preliminary diagnosis of the abnormal problem of the current equipment is made to obtain the preliminary diagnosis result. The operation state, prediction trend and preliminary diagnosis result of the equipment are displayed in a visual manner to the production operator and equipment maintenance participant in real time. However, the related art does not update and optimize the power of the park according to the actual demand of the power, which is not conducive to energy conservation and environmental protection. The operation state is not comprehensively judged according to the historical operation data, and the dynamic nature of fault response and the timeliness of maintenance report are lacking. SUMMARY

[0004] The technical problem solved by the present application is that the related art does not update and optimize the power of the park according to the actual demand of the power, which is not conducive to energy conservation and environmental protection, and the operation state is not comprehensively judged according to the historical operation data, and the dynamic nature of fault response and the timeliness of maintenance report are lacking.

[0005] To solve the above technical problems, the present application provides the following technical scheme: a smart park equipment maintenance optimization system based on a digital twin model, comprising a division module, a construction module and a regulation and control module.

[0006] The division module divides equipment groups according to equipment purposes.

[0007] The construction module builds a virtual operation unit, calculates a first correlation degree according to operation data of equipment in any virtual operation unit, simplifies the equipment in the virtual unit according to the first correlation degree, and performs power distribution in the simplified virtual operation unit according to a power consumption target to obtain a first power path;

[0008] The regulation module optimizes the power path according to the improved genetic algorithm to obtain a second power path, performs a first judgment according to real-time operation data of each device and the second power path, and performs a first operation according to a first judgment result.

[0009] As a preferred scheme of the intelligent park equipment maintenance optimization system based on the digital twin model, the device purposes include sensing purposes, network purposes, control and execution purposes, data storage purposes, security and alarm purposes, and transportation purposes, and the device purposes are obtained by summarizing device names through a language large model.

[0010] The sensing purpose devices include temperature and humidity sensors, air quality sensors, light sensors, intelligent water and electricity meters, cameras, access control devices, and vehicle recognition devices.

[0011] The network purpose devices include wired network devices, wireless network devices, and communication base stations.

[0012] The control and execution purpose devices include intelligent lighting devices, intelligent air conditioners and ventilation devices, and intelligent charging piles.

[0013] The data storage and processing purpose devices include servers and storage devices.

[0014] The security and alarm devices include fire detection devices, alarm speakers, and alarm lights.

[0015] The transportation purpose devices include patrol robots and delivery robots.

[0016] As a preferred scheme of the intelligent park equipment maintenance optimization system based on the digital twin model, the device groups include sensing groups, communication groups, execution groups, storage groups, security groups, and transportation groups, and the device purposes are obtained by summarizing device purposes through a language large model.

[0017] The virtual operation unit is constructed according to digital twin technology, and the construction logic of the virtual operation unit includes:

[0018] Obtain each device in any device group, first number the devices, the first number is a natural number, obtain the operation data of each device after the first numbering, the operation data is represented as the power consumption per hour, construct an operation matrix of the device according to the operation data corresponding to the device, and the operation matrix is represented as the operation data of the device within M days.

[0019] Obtain the size information and appearance information of the equipment, create a 3D model of each equipment through 3D modeling software, and add the corresponding operation matrix as the virtual operation parameter of the equipment;

[0020] After the setting is completed, name the file as a virtual operation unit.

[0021] As a preferred scheme of the intelligent park equipment maintenance optimization system based on the digital twin model, the calculation expression of the operation matrix is:

[0022]

[0023] Wherein, x i is the i-th equipment, a ij，1 ~a ij，24 is the power consumption of the i-th equipment at the 1st hour to the 24th hour of the jth day.

[0024] As a preferred scheme of the intelligent park equipment maintenance optimization system based on the digital twin model, the first condition is set, and the first condition is represented as the absolute value of the difference between the elements of the same row and the same column in the operation matrix of any two devices is less than the first value.

[0025] The calculation expression of the first correlation degree is:

[0026]

[0027] Wherein, P is the first correlation degree between any two devices, n1 is the number of elements in the operation matrix of any device meeting the first condition, and n2 is the total number of elements in any operation matrix.

[0028] As a preferred scheme of the intelligent park equipment maintenance optimization system based on the digital twin model, the logic of simplifying the equipment in the virtual operation unit according to the first correlation degree includes:

[0029] Select any device in any device group as a reference device, calculate the first correlation degree between the other devices in the same device group and the reference device, set the reference device as the first value, calculate the first product of the first value and each first correlation degree respectively, and set the first product as the second value of the corresponding device.

[0030] The logic of power distribution in the simplified virtual operation unit according to the power consumption target includes:

[0031] Set a total expected power consumption, which represents the total power consumption of the device group in one day. Calculate the sum of a first value and all second values, denoted as the first sum. Calculate a first ratio of the second values ​​to the first sum, or a second ratio of the first values ​​to the first sum. Set the first ratio as the power consumption coefficient of the corresponding device, and the second ratio as the power consumption coefficient of the reference device. Calculate the second product of the total expected power consumption and the power consumption coefficients, and set the second product as the expected power consumption of the corresponding device. Construct a first power consumption path based on the expected power consumption of each device. The expression for the first power consumption path is: E = {E1, E2, ... E...} Q}, where E is the first power path, is a set, and Q is the number of devices in the device group.

[0032] As a preferred embodiment of the smart park equipment maintenance and optimization system based on the digital twin model described in this invention, the following steps are taken: the operation matrix of any device in the equipment group is obtained, each row vector is represented as a power consumption path, each row vector is encoded, the encoding is an integer encoding, when the row vectors are the same, the corresponding row vector encoding is set to the same integer encoding, after encoding, K encodings are obtained, the population is initialized by a constrained random initialization method, the chromosome length is set to 24, the maximum value of the gene at each position of the chromosome is set to K, and the minimum value of the gene at each position of the chromosome is set to 1;

[0033] The individual preservation method is selected, and a single-point crossover method is adopted. A code position is randomly selected as the crossover point. The genes of the row vector where the crossover point is located are swapped with the code of the same position in the next row vector. Any code position is selected as a random mutation point. The code of the mutation point is updated to any integer between 1 and R. The fitness value is calculated. Whether the maximum number of generations has been reached is set as the termination condition. The process is iterated and a new running matrix is ​​generated until the fitness value reaches the termination condition, at which point the output of a new running matrix stops.

[0034] As a preferred embodiment of the smart park equipment maintenance and optimization system based on the digital twin model described in this invention, wherein: according to the logic of repeated power allocation of each new operation matrix, each new first power path is obtained, and each new first power path is set as a second power path.

[0035] As a preferred embodiment of the smart park equipment maintenance and optimization system based on digital twin model described in this invention, the system includes: making a first judgment based on the real-time operating data of each device and the second power path, and performing a first operation based on the result of the first judgment;

[0036] The logic of the first judgment includes:

[0037] Obtaining real-time running data of any device in the task group, the real-time running data being represented as real-time one-hour power consumption, obtaining a corresponding element in each second power path, denoted as a target element, calculating a first quotient value of the target element and 24, and obtaining a first difference value by subtracting the first quotient value from the real-time running data, taking an absolute value of the first difference value, setting a third value and a fourth value as thresholds of the absolute value of the first difference value, wherein the third value is less than the fourth value, performing a first comparison between the absolute value of the first difference value and the thresholds of the absolute value of the first difference value, and performing a first judgment according to a first comparison result.

[0038] As a preferred scheme of the smart park equipment maintenance optimization system based on the digital twin model, the absolute value of the first difference value is compared with the thresholds of the absolute value of the first difference value, and when the absolute value of the first difference value is distributed between the third value and the fourth value, the first judgment result is set as the device being in a normal state.

[0039] When the first difference value is less than or equal to the third value, the first judgment result is set as the device being in an optimized state, the target element is continuously reduced, and the first comparison result is obtained, and when all the first comparison results are that the absolute value of the first difference value is distributed between the third value and the fourth value, the reduction of the target element is stopped, and the target element at this time is updated as a new target element.

[0040] When the first difference value is greater than or equal to the fourth value, the next target element is jumped to, and the first comparison process is repeated, and when all the first comparison results are that the first difference value is greater than or equal to the fourth value, the first judgment result is set as the device being in an abnormal state, and when there is a first comparison result that the absolute value of the first difference value is distributed between the third value and the fourth value or the first difference value is less than or equal to the third value, the first judgment result is set as the device being in a normal state or the first judgment is set as the device being in an optimized state.

[0041] When the first judgment result is that the device is in an abnormal state, a first operation is set as sending a maintenance application, and the maintenance application includes a maintenance application time point and a maintenance device number, and when the first judgment result is that the device is in a normal state or the device is in an optimized state, the first operation is not performed.

[0042] The beneficial effects of the present application are: according to the device use, the device group is divided, so that the device management is more targeted. The construction module simplifies the devices in the virtual running unit by calculating the first correlation degree, reduces the complexity of the management object, improves the management efficiency, and the virtual running unit constructed by the digital twin technology can intuitively display the distribution and running state of the device. The operation and maintenance personnel can quickly locate the device position through the 3D visualization platform, view the detailed information of the device, and once an abnormality is found, the alarm and processing can be performed in time, which greatly improves the operation and maintenance efficiency. The improved genetic algorithm is used to optimize the power path to obtain the second power path, so that more accurate and efficient energy distribution is realized, the energy consumption is reduced, the energy consumption of the equipment is simulated and optimized, the system helps the park to realize energy saving and emission reduction, further reduces the operation cost, integrates the scattered equipment and data, breaks the information island between the traditional park systems, realizes the intelligent linkage between the systems, and improves the digital management ability of the park. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 The basic flowchart of the intelligent park equipment maintenance optimization system based on the digital twin model provided by an embodiment of the present application is shown. DETAILED DESCRIPTION

[0044] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments.

[0045] Embodiments, refer to Figure 1 For an embodiment of the present application, an intelligent park equipment maintenance optimization system based on a digital twin model is provided, which includes a division module, a construction module and a control module.

[0046] The division module divides the device group according to the device use;

[0047] The construction module constructs a virtual running unit, calculates a first correlation degree according to the running data of the devices in any virtual running unit, simplifies the devices in the virtual unit according to the first correlation degree, performs power distribution in the simplified virtual running unit according to the power consumption target, and obtains a first power path;

[0048] The control module optimizes the power path according to the improved genetic algorithm to obtain a second power path, performs a first judgment according to the real-time running data of each device and the second power path, and performs a first operation according to the first judgment result.

[0049] The application divides the device group according to the device use, so that the device management is more targeted. The construction module simplifies the devices in the virtual operation unit by calculating the first correlation degree, reduces the complexity of the management object, improves the management efficiency, and the virtual operation unit constructed by the digital twin technology can intuitively display the distribution and operation state of the device. The operation and maintenance personnel can quickly locate the device position through the 3D visualization platform, view the detailed information of the device, and once an abnormality is found, the system can alarm and process in time, greatly improving the operation and maintenance efficiency. The improved genetic algorithm is used to optimize the power path to obtain a second power path, so as to realize more accurate and efficient energy distribution, reduce energy consumption, simulate and optimize the energy consumption of the device, and help the park to realize energy saving and emission reduction, further reduce the operation cost, integrate the scattered devices and data, break the information island between the traditional park systems, realize the intelligent linkage between the systems, and improve the digital management ability of the park.

[0050] The device use includes sensing use, network use, control and execution use, data storage use, security and alarm use, and transportation use, the device use is obtained by summarizing the device name through the language large model;

[0051] The sensing use device includes a temperature and humidity sensor, an air quality sensor, an illumination sensor, a smart water meter, a camera, an access control device and a vehicle recognition device;

[0052] The network use device includes a wired network device, a wireless network device and a communication base station;

[0053] The control and execution use device includes a smart lighting device, a smart air conditioner and a ventilation device, and a smart charging pile;

[0054] The data storage and processing use device includes a server and a storage device;

[0055] The security and alarm device includes a fire detection device, an alarm horn and an alarm lamp;

[0056] The transportation use device includes a patrol robot and a delivery robot.

[0057] In a specific implementation, the device name is summarized and the purpose is divided by a language large model, and the precise classification of various devices in the park is realized. This classification method enables management personnel to develop more targeted maintenance strategies for devices with different purposes, avoiding the "one-size-fits-all" management mode, improving the fine level of device management, and by classifying devices by purpose, combining digital twin models and optimization algorithms, the system develops personalized energy consumption optimization strategies for devices with different purposes. For example, for devices with sensing purposes, energy consumption is reduced by adjusting the sampling frequency and data transmission method reasonably; for devices with control and execution purposes, the power and running time of the device are automatically adjusted according to actual needs to achieve energy saving and consumption reduction, based on device operation data and digital twin models, various devices can be monitored and analyzed in real time to discover potential faults in advance. For devices with sensing purposes, such as temperature and humidity sensors, air quality sensors, etc., the abnormal change of their data is monitored to determine whether the device has failed; for devices with control and execution purposes, such as intelligent lighting devices, intelligent air conditioners, etc., the service life of the device is predicted by analyzing its running state and energy consumption data. Early preventive maintenance effectively reduces the occurrence of device failures and prolongs the service life of the device.

[0058] The device group includes a sensing group, a communication group, an execution group, a storage group, a security group, and a transportation group, and the device purpose is summarized by a language large model and obtained;

[0059] The virtual running unit is constructed according to digital twin technology, and the construction logic of the virtual running unit includes:

[0060] Obtain each device in any device group, first number the device, the first number is a natural number, obtain the running data of each device after the first number, the running data is represented as the power consumption per hour, construct the running matrix of the device according to the running data corresponding to the device, the running matrix is represented as the running data of the device within M days;

[0061] Obtain the size information and appearance information of the device, create a 3D model of each device through a 3D modeling software, and add the corresponding running matrix as the virtual running parameter of the device;

[0062] After setting, name the file as a virtual running unit.

[0063] In a specific implementation, by running the matrix recording device in M days, the energy consumption of the equipment is analyzed in detail. For example, it is found that some equipment has a high energy consumption mode in a certain period of time, and the energy consumption is reduced by optimizing the operation strategy (such as adjusting the equipment running time or power), and the running data in the virtual running unit provides a basis for energy allocation. For example, in the construction module, according to the power consumption and importance of the equipment, the power is reasonably allocated to ensure the stable operation of the key equipment, and the overall energy consumption is optimized. The combination of 3D model and running data enables the manager to quickly locate the position and problem of the faulty equipment. For example, through the running matrix in the virtual running unit, it can be quickly judged whether the problem is caused by the equipment itself or the running environment, so as to improve the fault handling speed.

[0064] The calculation expression of the running matrix is:

[0065]

[0066] Wherein, x i is the i-th equipment, a ij，1 ~a ij，24 is the power consumption of the i-th equipment from the 1st hour to the 24th hour on the j-th day.

[0067] The first condition is set, and the first condition is represented as the absolute value of the difference between the elements in the same row and the same column of the running matrix of any two devices is less than the first value;

[0068] The calculation expression of the first correlation degree is:

[0069]

[0070] Wherein, P is the first correlation degree between any two devices, n1 is the number of elements in the running matrix of any device meeting the first condition, and n2 is the total number of elements in any running matrix.

[0071] In practical implementation, by calculating the first correlation degree P, the system can quantify the operational correlation between any two devices. This quantification method makes the correlation between devices more intuitive and operable, facilitating further analysis and optimization. The calculation of the first correlation degree is based on the device's operating matrix, i.e., the power consumption data of the device over a period of time. This analysis method based on actual operating data can truly reflect the operational correlation between devices, avoiding errors caused by subjective judgment. By setting a first condition (the absolute value of the difference between elements in the same row and column of the operating matrix of any two devices is less than a first value), the system quickly filters out devices with similar operating states. For these devices, similar maintenance strategies are adopted, thereby simplifying the management process and improving management efficiency. In the construction module, the system simplifies and optimizes devices based on the first correlation degree. For example, for devices with high correlation, energy is allocated to them as a whole, thereby improving energy utilization efficiency and avoiding energy waste caused by mutual influence between devices.

[0072] The logic for simplifying devices in a virtual runtime unit based on the first degree of association includes:

[0073] Select any device in any device group as a reference device, calculate the first correlation degree between other devices in the same device group and the reference device, set the reference device as the first value, calculate the first product of the first value and each first correlation degree, and set the first product as the second value of the corresponding device.

[0074] The logic for allocating power within the simplified virtual operating unit based on power consumption targets includes:

[0075] Set the total expected power consumption, which represents the total power consumption of the equipment group in one day. Calculate the sum of a first value and all second values, denoted as the first sum. Calculate the first ratio of the second value to the first sum, or the second ratio of the first value to the first sum. Set the first ratio as the power consumption coefficient of the corresponding equipment, and the second ratio as the power consumption coefficient of the reference equipment. Calculate the second product of the total expected power consumption and the power consumption coefficients, and set the second product as the expected power consumption of the corresponding equipment. Construct a first power consumption path based on the expected power consumption of each equipment. The expression for the first power consumption path is: E = {E1, E2, ... E...} Q}, where E is the first power path, is a set, and Q is the number of devices in the device group.

[0076] In a specific implementation, the simplified device group makes the power distribution more direct and efficient. By calculating the power coefficient of each device and distributing the expected power of each device according to the total expected power, the rationality and scientificity of power distribution are ensured, and the system is more accurate in power distribution by considering the correlation between devices. For example, more power is allocated to devices with high correlation to the reference device, while less power is allocated to devices with low correlation. This correlation-based power distribution method improves energy utilization efficiency and reduces energy waste. The logic of power distribution is based on the operation data and correlation of the devices, providing data-driven decision support for managers. This data-based decision-making method improves the scientificity and accuracy of decision-making.

[0077] The running matrix of any device in the device group is obtained, each row vector is represented as a power consumption path, each row vector is encoded, and the same row vector encoding is set to the same integer encoding when the row vectors are the same. After encoding, K encodings are obtained. The population is initialized by adopting a random initialization method with constraints. The chromosome length is set to 24, the maximum value of the gene at each position of the chromosome is set to K, and the minimum value of the gene at each position of the chromosome is set to 1.

[0078] The individual selection method is selected, and a single-point crossover method is used. A random encoding position is selected as the crossover point. The encoding of the same position of the row vector and the next row vector is exchanged. A random mutation point is selected, and the encoding of the mutation point is updated to any integer between 1 and R. The fitness value is calculated, and whether the maximum evolution generation is reached is set as the termination condition. New running matrices are iterated and generated until the fitness value reaches the termination condition, and the output of the new running matrix is stopped.

[0079] According to the logic of power distribution, each new first power path is obtained. Each new first power path is set as a second power path.

[0080] In a specific implementation, each row vector is represented as a power consumption path and is encoded as an integer, simplifying the representation of the power consumption path and facilitating the processing of the genetic algorithm. A random initialization method with constraints is adopted to ensure the diversity of the population while meeting the constraints of the problem. A random mutation point is selected to further increase the diversity of the population and avoid the algorithm falling into local optimum. The individual selection method is selected to retain excellent individuals and improve the convergence speed of the algorithm. The chromosome length is set to 24, the maximum value of the gene at each position of the chromosome is set to K, and the minimum value is set to 1, meeting the constraints of the problem. The improved genetic algorithm is used to optimize the power consumption path, realize data-driven decision-making, and improve the intelligent level of the park.

[0081] According to the real-time running data of each device and the second power path, a first judgment is made, and a first operation is performed according to the first judgment result;

[0082] The logic of the first judgment includes:

[0083] The real-time running data of any device in the task group is obtained, and the real-time running data is represented as the power consumption in real time for one hour. The corresponding element in each second power path is obtained, denoted as a target element. The first quotient value of the target element and 24 is calculated. The real-time running data is subtracted from the first quotient value to obtain a first difference value. The absolute value of the first difference value is taken. The third value and the fourth value are set as the threshold value of the absolute value of the first difference value, wherein the third value is less than the fourth value. The absolute value of the first difference value is compared with the threshold value of the absolute value of the first difference value, and the first judgment is made according to the first comparison result.

[0084] In specific implementation, the real-time running data of the device is obtained and compared with the target element in the second power path to monitor the running state of the device in real time. This real-time monitoring mechanism ensures that the device always runs on the expected power path, improves the stability of the device running, and dynamically adjusts the power distribution of the device according to the first judgment result. For example, if the absolute value of the first difference value exceeds the threshold value, it indicates that the real-time running data of the device deviates greatly from the expected power path, and the power distribution needs to be adjusted to ensure the normal operation of the device. The first quotient value of the target element and 24 is accurately calculated and compared with the real-time running data to accurately control the power consumption of the device. The third value and the fourth value are set as the threshold value of the absolute value of the first difference value, wherein the third value is less than the fourth value, and the adaptability of the system is improved.

[0085] The absolute value of the first difference value is compared with the threshold value of the absolute value of the first difference value. When the absolute value of the first difference value is distributed between the third value and the fourth value, the first judgment result is set as the device being in a normal state.

[0086] When the first difference value is less than or equal to the third value, the first judgment result is set as the device being in an optimized state. The target element is continuously reduced, and the first comparison result is obtained. When the first comparison result is that the absolute value of the first difference value is distributed between the third value and the fourth value, the reduction of the target element is stopped, and the target element at this time is updated as a new target element.

[0087] When the first difference is greater than or equal to the fourth value, the next target element is jumped to, and the first comparison process is repeated, when all the first comparison results are that the first difference is greater than or equal to the fourth value, the first judgment result is set as that the equipment is in an abnormal state, when there is a first comparison result that the absolute value of the first difference is distributed between the third value and the fourth value or the first difference is less than or equal to the third value, the first judgment result is set as that the equipment is in a normal state or the first judgment is set as that the equipment is in an optimized state.

[0088] When the first judgment result is that the equipment is in an abnormal state, the first operation is set as sending a maintenance application, the maintenance application includes a maintenance application time point and a maintenance equipment number, when the first judgment result is that the equipment is in a normal state or the equipment is in an optimized state, the first operation is not executed.

[0089] The application divides the equipment group according to the equipment use, so that the equipment management is more targeted. The construction module simplifies the equipment in the virtual operation unit by calculating the first correlation degree, reduces the complexity of the management object, improves the management efficiency, the virtual operation unit constructed by the digital twin technology can intuitively display the distribution and operation state of the equipment. The operation and maintenance personnel can quickly locate the equipment position through the 3D visualization platform, view the detailed information of the equipment, once the abnormality is found, the alarm and processing can be carried out in time, which greatly improves the operation and maintenance efficiency. The improved genetic algorithm is used to optimize the electric quantity path to obtain a second electric quantity path, so that more accurate and efficient energy distribution is realized, energy consumption is reduced, the system helps the park to realize energy saving and emission reduction, further reduces the operation cost, integrates the scattered equipment and data, breaks the information island between the traditional park systems, realizes the intelligent linkage between the systems, and improves the digital management ability of the park.

[0090] Those skilled in the art will appreciate that embodiments of the present application can be readily used as a method, a system or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (or computer- readable storage media) having computer-usable program code embodied in the medium. The medium may Figure 1 one or more functions specified in the flow or flows and / or blocks Figure 1 one or more functions specified in the flow or flows and / or blocks

[0091] It should be noted that the above-mentioned embodiments are only used to illustrate but not to limit the technical solutions of the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced, without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A system for smart park equipment maintenance optimization based on a digital twin model, characterized in that, The method comprises a division module, a construction module and a regulation module. The division module divides equipment groups according to equipment use; The construction module constructs a virtual running unit corresponding to the equipment group based on digital twinning technology, calculates a first correlation degree according to the running data of the equipment in any virtual running unit, simplifies the equipment in the virtual unit according to the first correlation degree, and performs power distribution in the simplified virtual running unit according to the power consumption target to obtain a first power path; The regulation module optimizes the power path according to an improved genetic algorithm to obtain a second power path, performs a first judgment according to the real-time running data of each device and the second power path, and performs a first operation according to the first judgment result; The logic of simplifying the equipment in the virtual running unit according to the first correlation degree comprises: selecting any device in any equipment group as a reference device, calculating the first correlation degree of other devices in the same equipment group with the reference device, setting the reference device as a first value, calculating the first product of the first value and each first correlation degree respectively, and setting the first product as the second value of the corresponding device; The logic of performing power distribution in the simplified virtual running unit according to the power consumption target comprises: setting a total expected electric quantity, the total expected electric quantity representing a total electric quantity consumed by the device group in a day, calculating a sum of the first values and the respective second values, denoted as a first sum, calculating a first ratio of the second values to the first sum or a second ratio of the first values to the first sum, setting the first ratio as an electric quantity coefficient of the corresponding device, setting the second ratio as an electric quantity coefficient of the reference device, calculating a second product of the total expected electric quantity and the electric quantity coefficient, setting the second product as an expected electric quantity of the corresponding device, and constructing a first electric quantity path according to the expected electric quantities of the respective devices, an expression of the first electric quantity path being: E = {E1, E2,... En}, wherein E is the first electric quantity path, n is a set, and Q is a number of devices in the device group. Q}, wherein E is the first electric quantity path, n is a set, and Q is a number of devices in the device group. obtaining the running matrix of any device in the equipment group, representing each row vector as a power consumption path, encoding each row vector, setting the same integer code for the corresponding row vector code when the row vectors are the same, obtaining K codes after encoding, initializing the population by adopting a random initialization with constraints, setting the chromosome length to 24, setting the maximum value of the gene at each position of the chromosome to K, and setting the minimum value of the gene at each position of the chromosome to 1; selecting individual preservation method and adopting single-point crossover method, randomly selecting a code position as a crossover point, exchanging the codes of the number of the row vector at the crossover point and the same position of the next row vector, selecting any code position as a random mutation point, updating the code of the mutation point to any integer between 1 and R, calculating the fitness value, setting whether the maximum evolution generation is reached as the termination condition, iterating and generating a new running matrix, and stopping outputting the new running matrix until the fitness value reaches the termination condition; repeating the logic of power distribution according to each new running matrix to obtain each new first power path, and setting the new first power paths as the second power paths.

2. The digital twin model based smart park device maintenance optimization system of claim 1, wherein: The equipment use includes sensing use, network use, control and execution use, data storage use, security and alarm use, and transportation use, which are obtained by summarizing the device names through a language large model; The sensing use devices include temperature and humidity sensors, air quality sensors, light sensors, intelligent water meters, cameras, access control devices and vehicle recognition devices; The network use devices include wired network devices, wireless network devices and communication base stations; The control and execution use devices include intelligent lighting devices, intelligent air conditioners and ventilation devices, and intelligent charging piles; The data storage and processing use devices include servers and storage devices; The security and alarm device includes a fire detection device, an alarm horn and an alarm lamp; The transportation device includes a patrol robot and a delivery robot. 3.The digital-twin-model-based smart park device maintenance optimization system of claim 1, wherein: The device group includes a perception group, a communication group, an execution group, a storage group, a security group and a transportation group, and the device purpose is summarized by the language large model and obtained; The virtual running unit is constructed according to the digital twin technology, and the construction logic of the virtual running unit includes: Each device in any device group is numbered, the first number is a natural number, the running data of each device after the first number is obtained, the running data is represented as the power consumption per hour, and the running matrix of the device is constructed according to the running data corresponding to the device, the running matrix is represented as the running data of the device in M days; The size information and appearance information of the device are obtained, the 3D model of each device is created by 3D modeling software, and the corresponding running matrix is added as the virtual running parameter of the device; After setting, the file is named as a virtual running unit.

4. The digital twin model based smart park device maintenance optimization system of claim 3, wherein: The calculation expression of the running matrix is: ； wherein x i is the ith device, a ij,1 ~a ij,24 is the power consumption of the jth day of the ith device.

5. The digital twin model based smart park device maintenance optimization system of claim 1, wherein: The first condition is set, which is represented as the absolute value of the difference between the elements in the same row and column of the running matrix of any two devices being less than the first value; The calculation expression of the first correlation degree is: ； Wherein, P is the first correlation degree between any two devices, n1 is the number of elements in the running matrix of any device meeting the first condition, and n2 is the total number of elements in any running matrix.

6. The digital twin model based smart park device maintenance optimization system of claim 1, wherein: First judgment is performed according to the real-time running data of each device and the second power path, and first operation is performed according to the first judgment result; The logic of the first judgment includes: The real-time running data of any device in the device group is obtained, the real-time running data is represented as the power consumption per hour, the corresponding element in each second power path is obtained, which is denoted as the target element, the first quotient value of the target element and 24 is calculated, the real-time running data and the first quotient value are subtracted to obtain the first difference value, the absolute value of the first difference value is taken, the third value and the fourth value are set as the threshold value of the absolute value of the first difference value, wherein the third value is less than the fourth value, the absolute value of the first difference value and the threshold value of the absolute value of the first difference value are compared, and the first judgment is performed according to the first comparison result.

7. The digital twin model based smart park device maintenance optimization system of claim 1, wherein: The first difference value and the threshold value of the absolute value of the first difference value are compared, when the first difference value is distributed between the third value and the fourth value, the first judgment result is set as the device being in normal state; When the first difference value is less than or equal to the third value, the first judgment result is set as the device being in an optimized state, the target element is continuously reduced, and the first comparison result is obtained, when the first comparison result is all that the first difference value is distributed between the third value and the fourth value, the reduction of the target element is stopped, and the target element at this time is updated as a new target element; When the first difference is greater than or equal to the fourth value, jump to the next target element, and repeat the first comparison process, when all the first comparison results are that the first difference is greater than or equal to the fourth value, set the first judgment result as the device being in an abnormal state, when there is a first comparison result that the absolute value of the first difference is distributed between the third value and the fourth value or the first difference is less than or equal to the third value, set the first judgment result as the device being in a normal state or set the first judgment as the device being in an optimized state; When the first judgment result is that the device is in an abnormal state, set the first operation as sending a maintenance application, the maintenance application including a maintenance application time point and a maintenance device number, when the first judgment result is that the device is in a normal state or the device is in an optimized state, do not perform the first operation.

Images