Agv trolley simulation model establishment method, system and medium based on digital twinning

Abstract

The application discloses an AGV trolley simulation model establishment method and system based on digital twinning, relates to the technical field of vehicle simulation models, and comprises the following steps: establishing an AGV trolley simulation model, and performing model performance evaluation on the AGV trolley simulation model; analyzing the driving path difference degree between historical driving records and driving simulation records; obtaining the historical driving records and the difference historical driving records in a transportation area, evaluating the simulation accuracy degree of a feature AGV trolley simulation model in the transportation area, and optimizing the feature AGV trolley simulation model in combination with the difference historical driving records to obtain a target AGV trolley simulation model; obtaining AGV trolley simulation condition data of the transportation area, inputting the AGV trolley simulation condition data into the target AGV trolley simulation model for simulation, obtaining target simulation records, and showing the target simulation records to a user.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of vehicle simulation models, in particular to an AGV trolley simulation model establishment method and system based on digital twinning and a medium. BACKGROUND

[0002] Digital twinning is a digital technology that creates a virtual model of a physical entity or system to monitor, analyze and optimize its performance in real time. The application of digital twinning technology in AGV trolley simulation model establishment has the following advantages: 1. High simulation, digital twinning can build a virtual environment similar to the actual AGV, which can improve the authenticity and reliability of the model; 2. Visual analysis, through digital twinning technology, the running state of AGV can be visualized in real time, which is convenient for operators to monitor and manage in real time; 3. Digital twinning technology can realize data synchronization between AGV virtual model and actual AGV, through real-time data flow, the virtual model can reflect the real-time state and operation of AGV in the real world.

[0003] At present, when using the established AGV trolley simulation model for simulation, the common method is to input the pre-set simulation parameters in the AGV trolley simulation model, and then simulate the AGV trolley. However, this method has a big problem. The environment for actually using the AGV trolley simulation model for simulation is different, and it is difficult to simulate and test various environments during the establishment of the AGV trolley simulation model, which leads to the fact that the user cannot verify the simulation results of the AGV trolley simulation model, and the output results of the AGV trolley simulation model may have a large gap with the actual results. SUMMARY

[0004] The purpose of the present application is to provide an AGV trolley simulation model establishment method and system based on digital twinning to solve the problems in the prior art.

[0005] To achieve the above purpose, the present application provides the following technical scheme: an AGV trolley simulation model establishment method based on digital twinning, the method comprising:

[0006] Step S100: Construct a simulation cloud platform, obtain device data of the AGV trolley, obtain characteristic environment data of the AGV trolley, establish an AGV trolley simulation model, perform model performance evaluation on the AGV trolley simulation model, and obtain a characteristic AGV trolley simulation model;

[0007] Step S200: Obtain the historical driving record of the AGV in the transportation area of the user, based on the historical driving record, use the feature AGV simulation model to simulate the driving state of the AGV in the transportation area, obtain the driving simulation record, analyze the driving path difference degree between the historical driving record and the driving simulation record, and obtain the difference historical driving record;

[0008] Step S300: Obtain the historical driving record and the difference historical driving record in the transportation area, evaluate the simulation accuracy of the feature AGV simulation model in the transportation area, and optimize the feature AGV simulation model based on the difference historical driving record to obtain a target AGV simulation model;

[0009] Step S400: Obtain the AGV simulation condition data of the transportation area, input the AGV simulation condition data into the target AGV simulation model for simulation, obtain a target simulation record, and display the target simulation record to the user.

[0010] Further, step S100 includes:

[0011] Step S101: Obtain the preset device data of the AGV from the cloud platform, the device data including data corresponding to each device parameter of the AGV;

[0012] Step S102: Obtain the preset feature environment data of the AGV from the cloud platform, the feature environment data including data corresponding to each environment parameter;

[0013] Step S103: Based on the device data and the feature environment data, use simulation software to establish an AGV simulation model of the AGV, and obtain marked simulation condition data for simulation from the cloud platform, wherein the marked simulation condition data includes data corresponding to each simulation condition parameter;

[0014] Step S104: Input the marked simulation condition data into the AGV simulation model, obtain the driving energy consumption data output by the AGV simulation model, the driving energy consumption data including the total distance L sum , average speed V △ , total device load H sum和 , and total device energy consumption E sum of the AGV after simulation by the AGV simulation model;

[0015] Step S105: Calculate the feature model performance ratio A of the AGV simulation model:

[0016]

[0017] Where g is the acceleration due to gravity; μ is the preset ground friction coefficient; C d is the preset air resistance coefficient; B sum is the total windward area of ​​the AGV;

[0018] Step S106: Perform model performance evaluation on the AGV trolley simulation model. The specific evaluation process is to obtain the preset first characteristic model performance ratio A′1 and second characteristic model performance ratio A′2 of the AGV trolley, where A′1>A′2>0. When the characteristic model performance ratio A′1>A>A′2, it is determined that the model performance of the AGV trolley simulation model meets the requirements, and the AGV trolley simulation model is recorded as the characteristic AGV trolley simulation model.

[0019] Furthermore, step S200 includes:

[0020] Step S201: Get the user who uses the characteristic AGV car simulation model, set the unit time, get the historical driving record of the AGV car in the user's transportation area, and get the driving path set D = {D1, D2, ..., D m}, where the two-dimensional coordinate points of the AGV at the 1st, 2nd, ..., mth unit time after driving in the historical driving record;

[0021] Step S202: Acquire characteristic simulation condition data of the AGV from the historical driving records. The characteristic simulation condition data includes data corresponding to various simulation condition parameters. Use the characteristic AGV simulation model to simulate the driving state of the AGV in the transportation area to obtain a driving simulation record. Obtain a simulated driving path F = {F1, F2, ..., F n}, where the two-dimensional coordinate points of the AGV in the driving simulation record at the 1st, 2nd, ..., nth unit time after the simulated driving;

[0022] Step S203: Analyze the degree of difference between the driving paths of the historical driving records and the driving simulation records. The specific analysis process is to construct a matrix W based on the driving path set D and the simulated driving path F, and set the first element in the matrix W to 0, where the element W in the αth row and βth column of the matrix W is α,β =L(D α ,F β ), where L(D α ,F β ) is the two-dimensional coordinate point D corresponding to the αth element in the driving path set D α , the two-dimensional coordinate point F corresponding to the βth element in the simulated driving path F β The Euclidean distance between them is as follows:

[0023]

[0024] wherein, x α , y α respectively represent the horizontal coordinate and the vertical coordinate of the two-dimensional coordinate point D α ; x β , y β respectively represent the horizontal coordinate and the vertical coordinate of the two-dimensional coordinate point F β ;

[0025] Step S204: defining a cumulative distance matrix Q, wherein the element Q i,j in the ith row and the ith column of the cumulative distance matrix Q is obtained as follows: Q i,j + min{Q i-1,j , Q i,j-1 , Q i-1,j-1}, wherein Q i-1,j represents the element in the (i-1)th row and the jth column of the cumulative distance matrix Q, Q i,j-1 represents the element in the ith row and the (j-1)th column of the cumulative distance matrix Q, Q i-1,j represents the element in the (i-1)th row and the jth column of the cumulative distance matrix Q, and Q i-1,j-1 represents the element in the (i-1)th row and the (j-1)th column of the cumulative distance matrix Q;

[0026] Step S205: obtaining the value corresponding to the last element Q m,n in the right lower corner of the cumulative distance matrix Q, and calculating the feature path difference value R between the historical driving record and the driving simulation record.

[0027]

[0028] When the feature path difference value R is greater than the preset feature path difference threshold value, it is determined that there is a difference in the driving path between the historical driving record and the driving simulation record of the AGV, and the historical driving record is recorded as the difference historical driving record of the transportation area.

[0029] Further, step S300 comprises:

[0030] Step S301: obtaining the total number U sum of the historical driving records in the transportation area that are simulated by the feature AGV simulation model, and obtaining the total number U' sum of the difference historical driving records in the transportation area;

[0031] Step S302: evaluating the simulation accuracy of the feature AGV simulation model in the transportation area, and the specific evaluation process is that the feature simulation accuracy score P of the feature AGV simulation model in the transportation area is calculated as follows: P = (U sum-U' sum ) / U sum ;

[0032] Step S303: When the feature AGV trolley simulation model is simulated in the feature simulation accuracy score P of the transportation area, it is determined that the feature AGV trolley simulation model is simulated in the simulation accuracy of the transportation area, and the feature AGV trolley simulation model is recorded as the target AGV trolley simulation model of the transportation area.

[0033] Step S304: When the feature simulation accuracy score P is less than the preset feature simulation accuracy score threshold, it is determined that the feature AGV trolley simulation model is not accurate in simulating the transportation area, and the feature AGV trolley simulation model is optimized, and the specific optimization process is:

[0034] Obtain the difference historical driving record in the transportation area, obtain the feature simulation condition data from the difference historical driving record, import the feature simulation condition data into the feature AGV trolley simulation model, obtain the feature simulation driving record of the difference historical driving record, and obtain the average value corresponding to each performance index from the feature simulation driving record.

[0035] Step S305: Obtain the average value corresponding to each performance index in the difference historical driving record, calculate the feature simulation error value K of the feature AGV trolley simulation model to the difference historical driving record:

[0036]

[0037] Wherein, S z is the average value corresponding to the zth performance index in the difference historical driving record; S′ z is the average value corresponding to the zth performance index in the feature simulation driving record; γ is the total number of each performance index in the difference historical driving record;

[0038] Step S306: Calculate the label simulation error ζ of the feature AGV trolley simulation model in the transportation area:

[0039]

[0040] Wherein, η represents the total number of the difference historical driving record in the transportation area; K ε represents the feature simulation error value of the feature AGV trolley simulation model to the εth difference historical driving record;

[0041] Step S307: Adjusting each model parameter of the feature AGV trolley simulation model, and calculating the adjusted feature AGV trolley simulation model, and the label simulation error in the transportation area until the label simulation error is less than the preset label simulation error threshold, determining that the feature AGV trolley simulation model is optimized, and recording the feature AGV trolley simulation model as the target AGV trolley simulation model of the transportation area;

[0042] In the above steps, the feature simulation accuracy score of the feature AGV trolley simulation model in the transportation area is calculated to determine whether the feature AGV trolley simulation model is used in the transportation area. Not only is the speed very fast, but the data used in the determination is most relevant to the transportation area, and the feature AGV trolley simulation model is most suitable for the transportation area, thereby providing data support for the feature AGV trolley simulation model below, which is beneficial to improve the simulation accuracy of the feature AGV trolley simulation model.

[0043] Further, step S400 includes:

[0044] Step S401: Obtaining the target AGV trolley simulation model of the transportation area, and obtaining the AGV trolley simulation condition data of the user in the transportation area as the driving scene. The AGV trolley simulation condition data includes data corresponding to each simulation condition parameter.

[0045] Step S402: Inputting the AGV trolley simulation condition data into the target AGV trolley simulation model to obtain a target simulation record, and displaying the target simulation record to the user through the cloud platform.

[0046] In order to better implement the above method, an AGV trolley simulation model establishment system is also proposed. The system includes a model evaluation module, a difference analysis module, a model optimization module, and a model simulation module.

[0047] The model evaluation module is used to evaluate the model performance of the AGV trolley simulation model to obtain the feature AGV trolley simulation model.

[0048] The difference analysis module is used to analyze the driving path difference between the historical driving record and the driving simulation record to obtain the difference historical driving record.

[0049] The model optimization module is used to optimize the feature AGV trolley simulation model to obtain the target AGV trolley simulation model.

[0050] The model simulation module is used to obtain the AGV trolley simulation condition data of the transportation area, input the AGV trolley simulation condition data into the target AGV trolley simulation model for simulation, obtain a target simulation record, and display the target simulation record to the user.

[0051] Further, the model evaluation module comprises a simulation model establishing unit and a model evaluation unit.

[0052] The simulation model establishing unit is configured to acquire equipment data of the AGV and acquire characteristic environment data of the AGV, and establish a simulation model of the AGV.

[0053] The model evaluation unit is configured to calculate a characteristic model performance ratio of the simulation model of the AGV, and perform model performance evaluation on the simulation model of the AGV according to the characteristic model performance ratio, to obtain a characteristic AGV simulation model.

[0054] Further, the difference analysis module comprises a characteristic path difference value unit and a difference analysis unit.

[0055] The characteristic path difference value unit is configured to acquire historical driving records of the AGV in the transportation area, acquire driving simulation records, and calculate a characteristic path difference value between the historical driving records and the driving simulation records.

[0056] The difference analysis unit is configured to analyze a driving path difference degree between the historical driving records and the driving simulation records according to the characteristic path difference value, to obtain a difference historical driving record.

[0057] Further, the model optimization module comprises a simulation accuracy evaluation unit and a model optimization unit.

[0058] The simulation accuracy evaluation unit is configured to evaluate a simulation accuracy degree of the characteristic AGV simulation model in the transportation area.

[0059] The model optimization unit is configured to optimize the characteristic AGV simulation model, to obtain a target AGV simulation model of the transportation area.

[0060] A medium, the medium stores computer instructions, when the computer instructions are executed by a processor, a simulation model establishing method of an AGV based on digital twinning can be realized.

[0061] Compared with the prior art, the beneficial effects of the present application are: the present application realizes the intelligent establishment of the AGV trolley simulation model, when using the AGV trolley simulation model to simulate in the non-transportation area, the simulation can be carried out according to the historical driving record of the AGV trolley in the actual transportation area, the driving simulation record is obtained, the driving path difference between the historical driving record and the driving simulation record is analyzed, thereby verifying whether the AGV trolley simulation model is accurate in the transportation area, and the AGV trolley simulation model can be optimized and adjusted, so that the established AGV trolley simulation model can be suitable for various environments and various AGV trolleys, and further, the accuracy of the AGV trolley simulation model is greatly improved. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 is a method flowchart of the AGV trolley simulation model establishment method, system and medium based on digital twinning of the present application;

[0063] Figure 2 is a module schematic diagram of the AGV trolley simulation model establishment method, system and medium based on digital twinning of the present application. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0065] Embodiment: as shown in the figure, the present application provides a technical solution, an AGV trolley simulation model establishment method based on digital twinning, the method comprises: Figure 1-Figure 2

[0066] Step S100: constructing a simulation cloud platform, obtaining device data of the AGV trolley, obtaining characteristic environment data of the AGV trolley, establishing an AGV trolley simulation model, performing model performance evaluation on the AGV trolley simulation model, and obtaining a characteristic AGV trolley simulation model;

[0067] Among them, step S100 includes:

[0068] Step S101: obtaining the preset device data of the AGV trolley from the cloud platform, the device data includes data corresponding to each device parameter of the AGV trolley;

[0069] For example, each device parameter includes the vehicle self-weight, the maximum driving speed of the AGV trolley, etc.

[0070] ​Step S102: Obtain the preset characteristic environment data of the AGV from the cloud platform, and the characteristic environment data includes data corresponding to each environmental parameter;

[0071] Step S103: Based on the device data and the characteristic environment data, use simulation software to establish an AGV simulation model of the AGV, and obtain marked simulation condition data from the cloud platform for simulation, wherein the marked simulation condition data includes data corresponding to each simulation condition parameter;

[0072] For example, the motor power, battery capacity, maximum acceleration, etc. of the AGV;

[0073] Step S104: Input the marked simulation condition data into the AGV simulation model, obtain the running energy consumption data output by the AGV simulation model, and the running energy consumption data includes the total distance L sum , average speed V △ , total load H sum和 of the AGV after simulation of the AGV simulation model, and total energy consumption E sum of the device;

[0074] Step S105: Calculate the characteristic model performance ratio A of the AGV simulation model:

[0075]

[0076] Wherein g is the acceleration of gravity; μ is the preset ground friction coefficient; C d is the preset air resistance coefficient; B sum is the total windward area of the AGV;

[0077] Step S106: Model performance evaluation is performed on the AGV simulation model, and the specific evaluation process is to obtain the preset first characteristic model performance ratio A'1 and second characteristic model performance ratio A'2 of the AGV, wherein A'1>A'2>0, when the characteristic model performance ratio A'1>A>A'2, it is determined that the model performance of the AGV simulation model meets the requirements, and the AGV simulation model is recorded as a characteristic AGV simulation model;

[0078] Step S200: Obtain the historical running record of the AGV in the transportation area of the user, use the characteristic AGV simulation model to simulate the running state of the AGV in the transportation area based on the historical running record, obtain the running simulation record, analyze the running path difference degree between the historical running record and the running simulation record, and obtain the difference historical running record;

[0079] Wherein, step S200 includes:

[0080] Step S201: Get the user who uses the characteristic AGV car simulation model, set the unit time, get the historical driving record of the AGV car in the user's transportation area, and get the driving path set D = {D1, D2, ..., D m}, where the two-dimensional coordinate points of the AGV at the 1st, 2nd, ..., mth unit time after driving in the historical driving record;

[0081] Step S202: Acquire characteristic simulation condition data of the AGV from the historical driving records. The characteristic simulation condition data includes data corresponding to various simulation condition parameters. Use the characteristic AGV simulation model to simulate the driving state of the AGV in the transportation area to obtain a driving simulation record. Obtain a simulated driving path F = {F1, F2, ..., F n}, where the two-dimensional coordinate points of the AGV in the driving simulation record at the 1st, 2nd, ..., nth unit time after the simulated driving;

[0082] Step S203: Analyze the degree of difference between the driving paths of the historical driving records and the driving simulation records. The specific analysis process is to construct a matrix W based on the driving path set D and the simulated driving path F, and set the first element in the matrix W to 0, where the element W in the αth row and βth column of the matrix W is α,β =L(D α ,F β ), where L(D α ,F β ) is the two-dimensional coordinate point D corresponding to the αth element in the driving path set D α , the two-dimensional coordinate point F corresponding to the βth element in the simulated driving path F β The Euclidean distance between them is as follows:

[0083]

[0084] Among them, x α 、y α Represented as two-dimensional coordinate points D α The horizontal and vertical coordinates of x β 、y β Represented as two-dimensional coordinate points F β The horizontal and vertical coordinates of

[0085] Step S204: define a cumulative distance matrix Q, wherein the element Q in the i-th row and i-th column of the cumulative distance matrix Q is obtained. i,j =W i,j +min{Q i-1,j ,Q i,j-1 ,Q i-1,j-1wherein Q i-1,j represents an element in the i-1th row and jth column of the accumulated distance matrix Q, Q i,j-1 represents an element in the ith row and j-1th column of the accumulated distance matrix Q, Q i-1,j represents an element in the i-1th row and jth column of the accumulated distance matrix Q, Q i-1,j-1 represents an element in the i-1th row and j-1th column of the accumulated distance matrix Q;

[0086] Step S205: obtaining the last element Q m,n corresponding numerical value, calculating the feature path difference value R between the historical driving record and the driving simulation record:

[0087]

[0088] When the feature path difference value R is greater than a preset feature path difference threshold value, it is determined that there is a difference in the driving path between the historical driving record and the driving simulation record, and the historical driving record is recorded as the difference historical driving record of the transportation area;

[0089] Step S300: obtaining the historical driving record and the difference historical driving record in the transportation area, evaluating the simulation accuracy of the feature AGV trolley simulation model in the transportation area, and optimizing the feature AGV trolley simulation model in combination with the difference historical driving record to obtain a target AGV trolley simulation model;

[0090] wherein step S300 comprises:

[0091] Step S301: obtaining the total number U of the historical driving records in the transportation area that are simulated by the feature AGV trolley simulation model sum , and obtaining the total number U' of the difference historical driving records in the transportation area sum ;

[0092] Step S302: evaluating the simulation accuracy of the feature AGV trolley simulation model in the transportation area, and the specific evaluation process is that the feature simulation accuracy score P of the feature AGV trolley simulation model in the transportation area is calculated as P=(U sum -U' sum ) / U sum ;

[0093] Step S303: when the feature simulation accuracy score P of the feature AGV trolley simulation model in the transportation area is greater than or equal to a preset feature simulation accuracy score threshold value, it is determined that the feature AGV trolley simulation model is accurate in simulating the transportation area, and the feature AGV trolley simulation model is recorded as the target AGV trolley simulation model of the transportation area;

[0094] Step S304: When the feature simulation accuracy score P is less than the preset feature simulation accuracy score threshold, it is determined that the feature AGV vehicle simulation model is not accurate in simulating the transportation area, and the feature AGV vehicle simulation model is optimized, and the specific optimization process is as follows:

[0095] The difference historical driving records in the transportation area are obtained, the feature simulation condition data are obtained from the difference historical driving records, the feature simulation condition data are imported into the feature AGV vehicle simulation model, the feature simulation driving records of the difference historical driving records are obtained, and the average values corresponding to each performance index are obtained from the feature simulation driving records;

[0096] For example, the performance indexes include the total length of the driving path, the completion time, etc.

[0097] Step S305: The average values corresponding to each performance index in the difference historical driving records are obtained, and the feature simulation error value K of the feature AGV vehicle simulation model to the difference historical driving records is calculated as follows:

[0098]

[0099] Wherein, S z is the average value corresponding to the zth performance index in the difference historical driving records; S' z is the average value corresponding to the zth performance index in the feature simulation driving records; γ is the total number of performance indexes in the difference historical driving records;

[0100] For example, γ is 4, S1 is 30; S2 is 20; S3 is 15; S4 is 10; S'1 is 33; S'2 is 23; S'3 is 17; S'4 is 11; the feature simulation error value K of the feature AGV vehicle simulation model to the difference historical driving records is calculated as follows:

[0101]

[0102] Step S306: The feature AGV vehicle simulation model is calculated, and the label simulation error ζ in the transportation area is calculated as follows:

[0103]

[0104] Wherein, η represents the total number of difference historical driving records in the transportation area; K ε represents the feature simulation error value of the feature AGV vehicle simulation model to the εth difference historical driving record;

[0105] Step S307: adjusting each model parameter of the characteristic AGV simulation model, and calculating the adjusted characteristic AGV simulation model, and the label simulation error in the transportation area until the label simulation error is less than the preset label simulation error threshold, determining that the characteristic AGV simulation model is optimized, and recording the characteristic AGV simulation model as the target AGV simulation model of the transportation area;

[0106] For example, each model parameter includes a PID control parameter and the like.

[0107] Step S400: obtaining AGV simulation condition data of the transportation area, inputting the AGV simulation condition data into the target AGV simulation model for simulation, obtaining a target simulation record, and displaying the target simulation record to the user.

[0108] The step S400 includes:

[0109] Step S401: obtaining the target AGV simulation model of the transportation area, and obtaining AGV simulation condition data of the user in the transportation area as a driving scene, the AGV simulation condition data including data corresponding to each simulation condition parameter.

[0110] Step S402: inputting the AGV simulation condition data into the target AGV simulation model to obtain a target simulation record, and displaying the target simulation record to the user through a cloud platform.

[0111] In order to better implement the above method, an AGV simulation model establishment system is also proposed, which includes a model evaluation module, a difference analysis module, a model optimization module, and a model simulation module.

[0112] The model evaluation module is configured to evaluate the performance of the AGV simulation model to obtain a characteristic AGV simulation model.

[0113] The difference analysis module is configured to analyze the driving path difference between the historical driving record and the driving simulation record to obtain a difference historical driving record.

[0114] The model optimization module is configured to optimize the characteristic AGV simulation model to obtain a target AGV simulation model.

[0115] The model simulation module is configured to obtain AGV simulation condition data of the transportation area, input the AGV simulation condition data into the target AGV simulation model for simulation, obtain a target simulation record, and display the target simulation record to the user.

[0116] The model evaluation module includes a simulation model establishment unit and a model evaluation unit.

[0117] The simulation model establishing unit is configured to acquire equipment data of the AGV and acquire characteristic environment data of the AGV, and establish a simulation model of the AGV;

[0118] The model evaluation unit is configured to calculate a characteristic model performance ratio of the simulation model of the AGV, and perform model performance evaluation on the simulation model of the AGV according to the characteristic model performance ratio, to obtain a characteristic AGV simulation model.

[0119] The difference analysis module includes a characteristic path difference value unit and a difference analysis unit.

[0120] The characteristic path difference value unit is configured to acquire historical driving records of the AGV in the transportation area, acquire driving simulation records, and calculate a characteristic path difference value between the historical driving records and the driving simulation records.

[0121] The difference analysis unit is configured to analyze a driving path difference degree between the historical driving records and the driving simulation records according to the characteristic path difference value, to obtain a difference historical driving record.

[0122] The model optimization module includes a simulation accuracy evaluation unit and a model optimization unit.

[0123] The simulation accuracy evaluation unit is configured to evaluate a simulation accuracy degree of the characteristic AGV simulation model in the transportation area.

[0124] The model optimization unit is configured to optimize the characteristic AGV simulation model, to obtain a target AGV simulation model of the transportation area.

[0125] A medium stores computer instructions, and the computer instructions are executed by a processor to implement an AGV simulation model establishing method based on digital twinning.

[0126] It is apparent to those skilled in the art that the present application is not limited to the details of the foregoing exemplary embodiments, and the present application can be implemented in other concrete forms without departing from the spirit or essential characteristics of the present application. Thus, the embodiments should be considered in all respects as illustrative and not restrictive, and the scope of the present application is defined by the appended claims rather than the foregoing description, and it is intended to include all changes falling within the meaning and range of equivalents of the elements of the claims. Any reference signs in the claims should not be considered as limiting the claims involved.

Claims

1. The method for establishing an AGV simulation model based on digital twins is characterized by: The method comprises: Step S100: constructing a simulation cloud platform, obtaining device data of the AGV, obtaining characteristic environmental data of the AGV, establishing an AGV simulation model, performing a model performance evaluation on the AGV simulation model, and obtaining a characteristic AGV simulation model; Step S200: Obtain historical driving records of the AGV in the user's transport area, simulate the driving state of the AGV in the transport area based on the historical driving records using the characteristic AGV simulation model to obtain a driving simulation record, analyze the degree of driving path difference between the historical driving records and the driving simulation record, and obtain a differential historical driving record; Step S300: Obtain historical driving records and differential historical driving records in the transport area, evaluate the simulation accuracy of the characteristic AGV simulation model in the transport area, and optimize the characteristic AGV simulation model based on the differential historical driving records to obtain a target AGV simulation model; Step S400: Acquire AGV trolley simulation condition data of the transport area, input the AGV trolley simulation condition data into the target AGV trolley simulation model for simulation, obtain target simulation records, and display the target simulation records to the user; The step S200 includes: Step S201: Obtain the user who uses the characteristic AGV car simulation model, set the unit time, obtain the historical driving record of the AGV car in the transportation area of ​​the user, and obtain the driving path set D = {D1, D2, ..., D m }, wherein the two-dimensional coordinate points of the AGV at the 1st, 2nd, ..., mth unit time after the driving in the historical driving record; Step S202: Acquire characteristic simulation condition data of the AGV from the historical driving record, wherein the characteristic simulation condition data includes data corresponding to various simulation condition parameters, and use the characteristic AGV simulation model to simulate the driving state of the AGV in the transportation area to obtain a driving simulation record, and obtain a simulated driving path F={F1, F2, ..., F n }, wherein the two-dimensional coordinate points of the AGV in the driving simulation record at the 1st, 2nd, ..., nth unit time after the simulated driving; Step S203: Analyze the degree of difference between the driving paths of the historical driving record and the driving simulation record. The specific analysis process is to construct a matrix W based on the driving path set D and the simulated driving path F, and set the first element in the matrix W to 0, wherein the element W in the αth row and βth column of the matrix W is α,β =L(D α ,F β ), where L(D α ,F β ) is the two-dimensional coordinate point D corresponding to the αth element in the driving path set D α , the two-dimensional coordinate point F corresponding to the βth element in the simulated driving path F β The Euclidean distance between them is as follows: ， Step S204: define a cumulative distance matrix Q, wherein the element Q in the i-th row and j-th column of the cumulative distance matrix Q is obtained. i,j =W i,j +min{Q i-1,j ,Q i,j-1 ,Q i-1,j-1 }; Step S205: Obtain the last element Q in the lower right corner of the cumulative distance matrix Q m,n The corresponding values ​​are used to calculate the characteristic path difference value R between the historical driving record and the driving simulation record: ， When the characteristic path difference value R is greater than a preset characteristic path difference threshold, it is determined that there is a difference in the driving path of the AGV between the historical driving record and the driving simulation record, and the historical driving record is recorded as a difference historical driving record of the transportation area.

2. The method for establishing an AGV simulation model based on digital twin according to claim 1 is characterized in that: The step S100 includes: Step S101: Obtaining preset AGV equipment data from the cloud platform, wherein the equipment data includes data corresponding to various equipment parameters of the AGV; Step S102: Obtaining preset characteristic environmental data of the AGV from the cloud platform, wherein the characteristic environmental data includes data corresponding to various environmental parameters; Step S103: Based on the device data and characteristic environment data, using simulation software, establish an AGV simulation model of the AGV, and obtain labeled simulation condition data for simulation from the cloud platform, wherein the labeled simulation condition data includes data corresponding to various simulation condition parameters; Step S104: Input the marked simulation condition data into the AGV car simulation model to obtain the driving energy consumption data output by the AGV car simulation model. The driving energy consumption data includes the total distance L traveled by the AGV car after the AGV car simulation model performs simulation. sum , average rate V △ , total equipment load H sum and total energy consumption of the equipment E sum ; Step S105: Calculate the characteristic model performance ratio A of the AGV simulation model: ， Where g is the acceleration due to gravity; μ is the preset ground friction coefficient; C d is the preset air resistance coefficient; B sum is the total windward area of ​​the AGV; Step S106: Perform model performance evaluation on the AGV trolley simulation model. The specific evaluation process is to obtain the preset first characteristic model performance ratio A´1 and second characteristic model performance ratio A´2 of the AGV trolley, where A´1>A´2>0. When the characteristic model performance ratio A´1>A>A´2, it is determined that the model performance of the AGV trolley simulation model meets the requirements, and the AGV trolley simulation model is recorded as the characteristic AGV trolley simulation model.

3. The method for establishing an AGV simulation model based on digital twin according to claim 2 is characterized in that: The step S300 includes: Step S301: Obtain the total number U of several historical driving records simulated by the characteristic AGV car simulation model in the transportation area sum , obtain the total number U of each difference historical driving record in the transportation area sum ; Step S302: Evaluate the simulation accuracy of the characteristic AGV car simulation model in the transportation area. The specific evaluation process is to calculate the characteristic simulation accuracy score P of the characteristic AGV car simulation model in the transportation area = (U sum -U´ sum ) / U sum ; Step S303: When the feature simulation accuracy score P of the characteristic AGV car simulation model in the transportation area is greater than or equal to a preset feature simulation accuracy score threshold, it is determined that the feature AGV car simulation model accurately simulates the transportation area, and the feature AGV car simulation model is recorded as the target AGV car simulation model of the transportation area; Step S304: When the feature simulation accuracy score P is less than a preset feature simulation accuracy score threshold, it is determined that the feature AGV car simulation model is inaccurate in simulating the transport area, and the feature AGV car simulation model is optimized. The specific optimization process is as follows: Obtaining difference historical driving records in the transportation area, obtaining characteristic simulation condition data from the difference historical driving records, importing the characteristic simulation condition data into the characteristic AGV simulation model, obtaining characteristic simulation driving records of the difference historical driving records, and obtaining average values ​​corresponding to various performance indicators from the characteristic simulation driving records; Step S305: Obtain the average value corresponding to each performance indicator in the difference historical driving record, calculate the characteristic AGV car simulation model, and calculate the characteristic simulation error value K of the difference historical driving record: ， Among them, S z is the average value corresponding to the zth performance indicator in the difference historical driving record; S´ z is the average value corresponding to the zth performance indicator in the characteristic simulation driving record; γ is the total number of performance indicators in the difference historical driving record; Step S306: Calculate the marking simulation error ζ of the characteristic AGV simulation model in the transportation area: ， Wherein, η represents the total number of different historical driving records in the transportation area; K ε It is represented by the characteristic simulation error value of the characteristic AGV car simulation model for the εth difference historical driving record; Step S307: Adjust the various model parameters of the characteristic AGV trolley simulation model, and calculate the marking simulation error of the adjusted characteristic AGV trolley simulation model in the transportation area, until the marking simulation error is less than the preset marking simulation error threshold, and determine that the optimization of the characteristic AGV trolley simulation model is completed, and record the characteristic AGV trolley simulation model as the target AGV trolley simulation model of the transportation area.

4. The method for establishing an AGV simulation model based on digital twin according to claim 3 is characterized in that: The step S400 includes: Step S401: obtaining a target AGV simulation model of the transport area, and obtaining AGV simulation condition data of the user using the transport area as a driving scene, wherein the AGV simulation condition data includes data corresponding to various simulation condition parameters; Step S402: Input the AGV car simulation condition data into the target AGV car simulation model to obtain a target simulation record, and display the target simulation record to the user through the cloud platform.

5. An AGV trolley simulation model establishment system, used to execute the AGV trolley simulation model establishment method based on digital twin according to any one of claims 1 to 4, characterized in that: The system includes a model evaluation module, a difference analysis module, a model optimization module, and a model simulation module; The model evaluation module is used to perform model performance evaluation on the AGV car simulation model to obtain a characteristic AGV car simulation model; The difference analysis module is used to analyze the degree of difference in the driving path between the historical driving record and the driving simulation record to obtain a difference historical driving record; The model optimization module is used to optimize the characteristic AGV car simulation model to obtain a target AGV car simulation model; The model simulation module is used to obtain the AGV car simulation condition data of the transportation area, input the AGV car simulation condition data into the target AGV car simulation model for simulation, obtain the target simulation record, and display the target simulation record to the user.

6. The AGV simulation model building system according to claim 5, characterized in that: The model evaluation module includes a simulation model establishment unit and a model evaluation unit; The simulation model building unit is used to acquire the equipment data of the AGV car, obtain the characteristic environment data of the AGV car, and build an AGV car simulation model; The model evaluation unit is used to calculate the characteristic model performance ratio of the AGV trolley simulation model, and perform model performance evaluation on the AGV trolley simulation model according to the characteristic model performance ratio to obtain a characteristic AGV trolley simulation model.

7. The AGV simulation model building system according to claim 5, characterized in that: The difference analysis module includes a characteristic path difference value unit and a difference analysis unit; The characteristic path difference value unit is used to obtain the historical driving records of the AGV in the transportation area, obtain the driving simulation records, and calculate the characteristic path difference value between the historical driving records and the driving simulation records; The difference analysis unit is used to analyze the degree of difference in the driving paths between the historical driving record and the driving simulation record according to the characteristic path difference value, and obtain a difference historical driving record.

8. The AGV simulation model building system according to claim 5, characterized in that: The model optimization module includes a simulation accuracy evaluation unit and a model optimization unit; The simulation accuracy evaluation unit is used to evaluate the simulation accuracy of the characteristic AGV simulation model in the transportation area; The model optimization unit is used to optimize the characteristic AGV trolley simulation model to obtain the target AGV trolley simulation model of the transportation area.

9. A medium, characterized in that The medium stores computer instructions, and when the computer instructions are executed by the processor, the method for establishing an AGV vehicle simulation model based on digital twins according to any one of claims 1 to 4 can be implemented.

Images