Dynamic optimization and release method for digital twin model

Abstract

The invention relates to the technical field of data processing, in particular to a dynamic optimization and release method for a digital twin model, and the method comprises the steps: collecting real-time data; determining a temporary grid; judging a simulation grid; calculating a prediction optimization index; simulating to form an optimization index; adjusting the side length of the grid; and issuing the optimized twinborn model. According to the method, the real-time density of the personnel in each grid in the industrial area, the real-time average moving speed of the personnel and the operation rate of equipment associated with the personnel are dynamically acquired and analyzed, so that high-precision and self-adaptive optimization of the digital twin model is realized. The personnel density reflects the operation concentration degree in a certain area, the average moving speed reveals the intensity of personnel mobility, and the personnel density and the average moving speed jointly form space and time distribution characteristics of personnel behaviors. The problems of slow response speed and low processing precision for dynamic change due to excessive dependence on historical data and a single optimization strategy are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method for dynamic optimization and publishing of digital twin models. Background Art

[0002] With the continuous improvement of industrial automation and intelligence levels, the traditional production management mode can no longer meet the real-time optimization requirements in complex dynamic environments. As a virtualization technical means, digital twin technology can achieve precise simulation, monitoring, and prediction of complex production processes by establishing real-time interaction between virtual models and physical systems, and can help enterprises optimize resource allocation and improve production efficiency. Therefore, how to construct an efficient digital twin model and achieve its dynamic optimization and accurate publishing has become an important direction for the development of modern intelligent manufacturing and industrial Internet of Things.

[0003] The patent document with the publication number CN118657254A discloses an energy efficiency optimization digital twin system for industrial parks based on large model algorithms. The system includes: a data acquisition module, a digital twin module, an energy efficiency calculation module, and a simulation optimization module; the data acquisition module is used to collect operation information in the park, the digital twin module constructs a digital twin model according to the operation information, the energy efficiency calculation module is used to calculate the energy efficiency information of the operation, and the simulation optimization module is used to simulate the operation of the digital twin module and give process optimization suggestions; the data acquisition module includes an operation monitoring unit, an information collection unit, and a data sorting unit. The operation monitoring unit is used to monitor the operation status of each operation, the information collection unit is used to collect data information during the operation of the operation, and the data sorting unit is used to classify and sort the collected data information; the digital twin module includes an operation basic unit, an operation linkage unit, a data storage unit, and a real-time display unit. The operation basic unit is used to record the basic data of each operation, the operation linkage unit is used to record the linkage relationship and linkage data between operations, the data storage unit is used to save the collected data information, and the real-time display unit is used to display real-time operation information; the energy efficiency calculation module includes an energy consumption calculation unit, an efficiency calculation unit, and a comprehensive calculation unit. The energy consumption calculation unit is used to calculate the energy consumption information of the operation, the efficiency calculation unit is used to calculate the efficiency information of the operation, and the comprehensive calculation unit is used to calculate the energy efficiency information of the entire park; the simulation optimization module includes a model construction unit, a simulation operation unit, and an optimization output unit. The model construction unit constructs a data conversion model based on historical data, the simulation operation unit simulates the operation of the operation under different strategies based on real-time data and the data conversion model, and the optimization output unit is used to select and output the strategy with the best energy efficiency as the optimization plan.

[0004] It can be seen that the digital twin system for optimizing the energy efficiency of industrial parks based on large model algorithms has the following problems: This system overly relies on model construction and simulation operation units based on historical data, resulting in the model not being responsive enough when facing a rapidly changing production environment; This system mainly targets unified energy efficiency optimization within the park, assuming that the operations or equipment within the park are relatively single. However, in actual industrial parks, there are multiple different types of equipment, different kinds of operations, and diverse operating environments. In such a heterogeneous system, a single optimization strategy is difficult to adapt to the needs and characteristics of different equipment. Summary of the Invention

[0005] To this end, the present invention provides a method for dynamically optimizing and publishing a digital twin model, which is used to overcome the problems of slow response speed and low processing accuracy in dealing with dynamic changes due to over-reliance on historical data and a single optimization strategy in the prior art by introducing real-time data feedback and an adaptive optimization mechanism.

[0006] To achieve the above object, the present invention provides a method for dynamically optimizing and publishing a digital twin model, including: Collect the real-time operation rates of several devices in each to-be-simulated grid within an industrial area constructed based on a preset grid side length, the real-time density of targets in each to-be-simulated grid, and the real-time average moving speed. Determine several temporary grids according to the real-time operation rates within a preset determination duration and a preset operation rate fluctuation threshold. Screen out several simulation grids according to the real-time density between any two adjacent temporary grids within a preset determination duration. Calculate a prediction optimization index according to the real-time operation rates and the real-time density in each simulation grid. Use a preset digital twin model to simulate all the to-be-simulated grids and generate a simulation optimization index. Adjust the preset grid side length according to the prediction optimization index and the simulation optimization index to form an adjusted grid side length. Use the real-time operation rates, the real-time density, and the real-time average moving speed of all the simulation grids determined based on the adjusted grid side length to train the preset digital twin model to form an optimized twin model. Correct the preset operation rate fluctuation threshold according to the simulation optimization index re-simulated based on the optimized twin model within a preset correction duration to form a corrected operation rate fluctuation threshold. Publish the optimized twin model re-formed based on the corrected operation rate fluctuation threshold.

[0007] Furthermore, the process of determining several temporary grids according to the real-time operation rates within a preset determination duration and a preset operation rate fluctuation threshold includes: Calculate the standard deviation of the real-time operation rate to form an operation rate fluctuation value; When the operation rate fluctuation value is greater than the preset operation rate fluctuation threshold, determine the to-be-simulated grid as a temporary grid to form a number of temporary grids.

[0008] Further, the process of screening a number of simulation grids according to the real-time density in any two adjacent temporary grids within a preset determination duration includes: Calculate the standard deviation of the real-time density of each temporary grid to form a density fluctuation value; Screen a number of simulation grids according to the density fluctuation values of any two adjacent temporary grids.

[0009] Further, the process of screening a number of simulation grids according to the density fluctuation values of any two adjacent temporary grids includes: Calculate the relative deviation of the density fluctuation values of any two adjacent temporary grids to form a fluctuation consistency; When the fluctuation consistency is greater than the preset consistency threshold, determine that the temporary grids are all the simulation grids to form a number of simulation grids.

[0010] Further, the process of calculating a prediction optimization index according to the real-time operation rate and the real-time density in each simulation grid includes: Perform a normalization calculation on the real-time density of all the simulation grids to form a normalized density; Perform a normalization calculation on the real-time operation rate of all the simulation grids to form a normalized operation rate; Perform a weighted sum according to the preset density weight, the preset air flow rate weight, the normalized density, and the normalized operation rate to form a prediction optimization index.

[0011] Further, the process of using a preset digital twin model to simulate all the to-be-simulated grids and generate a simulation optimization index includes: Obtain the real-time operation rate and the real-time density of all the to-be-simulated grids and input them into the preset digital twin model to form a simulation optimization index.

[0012] Further, the process of adjusting the preset grid side length according to the prediction optimization index and the simulation optimization index includes: Calculate the relative deviation of the prediction optimization index and the simulation optimization index to form an index deviation; Adjust the preset grid side length according to the index deviation and the real-time average moving speed.

[0013] Further, the process of adjusting the preset grid side length according to the exponential deviation and the real-time average moving speed includes: When the deviation comparison result shows that the exponential deviation is greater than the preset deviation threshold, increase the preset grid side length according to the relative deviation between the exponential deviation and the preset deviation threshold and the preset adjustment coefficient to form a temporary grid side length; Adjust the temporary grid side length according to all the temporary grid side lengths within the preset adjustment duration and the real-time average moving speed to form an adjusted grid side length.

[0014] Further, the process of adjusting the temporary grid side length according to all the temporary grid side lengths within the preset adjustment duration and the real-time average moving speed to form an adjusted grid side length includes: Calculate the standard deviation of the formation times of all the temporary grid side lengths and normalize it to form a normalized number fluctuation value; Calculate the standard deviation of all the real-time average moving speeds and normalize it to form a normalized speed fluctuation value; Calculate the correlation coefficient between the normalized number fluctuation value and the normalized speed fluctuation value to form a change synchronization degree; When the change synchronization degree is less than the preset standard synchronization degree, increase the temporary grid side length according to the relative deviation between the preset standard synchronization degree and the change synchronization degree and the preset adjustment coefficient to form an adjusted grid side length.

[0015] Further, the process of correcting the preset operation rate fluctuation threshold according to the simulated optimization index re-simulated based on the optimized twin model within the preset correction duration to form a corrected operation rate fluctuation threshold includes: Calculate the standard deviation of the simulated optimization index to form an optimization index fluctuation value; When the optimization index fluctuation value is greater than the preset index fluctuation threshold, decrease the preset operation rate fluctuation threshold according to the relative deviation between the optimization index fluctuation value and the preset index fluctuation threshold and the preset correction coefficient to form a corrected operation rate fluctuation threshold.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows. By dynamically collecting and analyzing the real-time density of personnel in each grid within the industrial area, the real-time average moving speed of personnel, and the associated equipment operation rate, high-precision and self-adaptive optimization of the digital twin model are achieved. The personnel density reflects the degree of operation concentration in a certain area, and the average moving speed reveals the strength of personnel mobility. Together, they constitute the spatial and temporal distribution characteristics of personnel behavior. As an external influencing factor of personnel behavior, the fluctuation of the equipment operation rate reflects the change in the state of personnel operating the equipment. Combining the density and speed changes can comprehensively determine the work intensity and rhythm of personnel in the area. Introducing the "change synchronization degree" as a natural feedback adjustment mechanism, by measuring the consistency between personnel mobility and regional adjustment behavior, it guides the dynamic update of the grid structure, thereby enabling the twin model to have higher stability and response efficiency in complex environments, significantly improving the intelligent level of personnel behavior simulation, efficiency evaluation, and spatial resource allocation, and effectively solving the problems of slow response speed and low processing accuracy caused by over-reliance on historical data and single optimization strategies.

[0017] Furthermore, the calculation of the real-time operation rate and the fluctuation value directly reflects the operation stability of the equipment or area. Therefore, the temporary grid identification when the fluctuation value exceeds the threshold helps to discover potential operation problems or abnormal areas, enhances the dynamic response ability of the model, can quickly adapt to the changing operation state, and perform targeted optimization and adjustment to improve the overall stability and efficiency of the system.

[0018] Furthermore, by comparing the density fluctuation values of adjacent temporary grids, the division of dynamic areas is achieved, effectively improving the rationality and accuracy of the simulation grid. The relationship between the density fluctuation value and the simulation grid reflects the stability and consistency of the personnel distribution. When the density fluctuation is large, it means there are strong dynamic changes in the area and more refined division is needed; when the fluctuation is small, it indicates that the distribution of personnel or objects in the area is relatively uniform and can be combined into a larger simulation grid. It can perform precise grid division according to real-time changes, which helps to optimize the model simulation effect, improve the calculation efficiency, and ensure the reliability of the simulation results.

[0019] Furthermore, by using the fluctuation consistency to judge the merging of adjacent temporary grids, the rationality and accuracy of the simulation grid are ensured. The relationship between the fluctuation consistency and the density fluctuation value can reflect the stability of the personnel distribution in the area. When the density fluctuations of two adjacent grids are very similar, it means that the dynamic changes in these two grids are almost the same. Merging them into one simulation grid can simplify the model calculation without affecting the accuracy. It can dynamically adjust the division of the simulation grid according to the actual density fluctuation characteristics, optimize the simulation efficiency and accuracy, and avoid unnecessary over-division, thereby improving the overall simulation performance.

[0020] Furthermore, by comprehensively considering the impact of real-time density and operation rate factors on the optimization results, the prediction accuracy and operation feasibility can be improved. Through normalization and weighting processes, the impacts of various parameters are quantified and standardized, making the optimization index more in line with actual requirements, effectively evaluating the performance of different simulation grids, and thus providing a reliable basis for subsequent grid adjustment and optimization decisions.

[0021] Furthermore, by introducing the real-time average moving speed as the third parameter, the model can more comprehensively reflect the dynamic behavior of personnel, i.e., the moving speed of personnel. This additional dynamic parameter enables the model not to stay at the analysis of static data but to further optimize the system simulation by combining the personnel flow situation. Specifically, the introduction of the moving speed enables the model to identify the flow trend of personnel between different grids, further optimizing the matching between personnel density and equipment operation efficiency, and avoiding problems such as congestion or low efficiency caused by excessive personnel density or heavy equipment load.

[0022] Furthermore, by introducing the comparison between the prediction optimization index and the simulation optimization index, the deviation of the current grid design in practical applications can be effectively evaluated. The index deviation reflects the difference between the prediction result and the actual simulation result, which is the basis for adjusting the grid side length, provides the gap between the model prediction and the actual situation, helps identify the optimization space, and by combining the real-time average moving speed, the impact of personnel mobility is considered, helping to adjust the grid side length to adapt to the personnel dynamics in actual operation.

[0023] Furthermore, by combining the use of the index deviation and the real-time average moving speed, a more refined grid adjustment strategy is achieved. The index deviation provides a measure of the difference between the model prediction and the actual situation, making the grid adjustment more targeted. When the index deviation is greater than the preset deviation threshold, the grid side length is adjusted to quickly respond to the deviation between the model and reality, thereby improving the model accuracy and precision. Introducing the real-time average moving speed as a dynamic factor, considering the impact of personnel flow on the grid distribution, can further optimize the grid adjustment to ensure that it can adapt to the actual personnel flow situation, avoiding overly large or small grid divisions, and thus enhancing the efficiency of resource scheduling.

[0024] Furthermore, by normalizing the formation times of the temporary grid side lengths and the volatility of the real-time average movement speed, and measuring their change synchronization degree, the relationship between the grid side lengths and the personnel flow can be captured more accurately. This process ensures the adaptability of the model in the face of dynamic changes, avoiding incorrect predictions caused by excessive fluctuations or desynchronization. The normalized fluctuation value of the formation times reflects the changes in the grid side lengths, showing the impact of system state or environmental changes on grid division. The normalized fluctuation value of the movement speed reflects the fluctuations in the dynamic movement speed of personnel, revealing the impact of personnel mobility on the entire area. The change synchronization degree provides the correlation between the two, measuring the matching degree between personnel mobility and grid division, ensuring that in practical applications, grid adjustments and personnel dynamics can change synchronously.

[0025] Furthermore, by calculating the fluctuation value of the simulation optimization index and comparing it with a preset threshold, the fluctuation threshold of the operation rate can be dynamically adjusted according to the deviation, ensuring that the model can more precisely adapt to the changes in the actual operating environment. The self-correction mechanism improves the flexibility and adaptability of the system, avoiding errors caused by overly strict or loose threshold settings, ensuring the efficient and accurate operation of the model in different scenarios, and thus enhancing the overall optimization ability and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a flowchart of the method for dynamic optimization and release of the digital twin model in this embodiment; Figure 2 is a decision logic diagram for determining the temporary grid in this embodiment; Figure 3 is a decision logic diagram for determining the simulation grid in this embodiment; Figure 4 is a decision logic diagram for forming the adjusted grid side lengths in this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] In order to make the objectives and advantages of the present invention clearer, the present invention will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0028] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present invention and do not limit the protection scope of the present invention.

[0029] Please refer to Figure 1 as shown, which is a flowchart of the method for dynamic optimization and release of the digital twin model in this embodiment; This embodiment provides a method for dynamic optimization and release of a digital twin model, including: Collect the real-time operation rate of several devices in each to-be-simulated grid within the industrial area constructed based on the preset grid side length, the real-time density of targets in each to-be-simulated grid, and the real-time average moving speed. Determine several temporary grids according to the real-time operation rate within the preset determination duration and the preset operation rate fluctuation threshold. Screen out several simulation grids according to the real-time density within any two adjacent temporary grids within the preset determination duration. Calculate the prediction optimization index according to the real-time operation rate and the real-time density within each simulation grid. Use the preset digital twin model to simulate all the to-be-simulated grids and generate a simulation optimization index. Adjust the preset grid side length according to the prediction optimization index and the simulation optimization index to form an adjusted grid side length. Use the real-time operation rate, the real-time density, and the real-time average moving speed of all the simulation grids determined based on the adjusted grid side length to train the preset digital twin model to form an optimized twin model. Correct the preset operation rate fluctuation threshold according to the simulation optimization index re-simulated based on the optimized twin model within the preset correction duration to form a corrected operation rate fluctuation threshold. Release the optimized twin model reformed based on the corrected operation rate fluctuation threshold.

[0030] The process of releasing the optimized twin model reformed based on the corrected operation rate fluctuation threshold includes inputting the adjusted and corrected model parameters and related data into the system, updating the core algorithm of the digital twin model, and releasing the updated optimized model through the system interface or platform to ensure that each relevant module can obtain and apply the new optimization rules and operation thresholds in a timely manner. After release, the new optimized model will be automatically confirmed and applied to the actual operation, and the system performance will be monitored and fed back in real time to continuously optimize the operation efficiency and safety.

[0031] The real-time operation rate of the device refers to the ratio of the number of devices actually in the running state in a certain grid at a specific moment or time window to the total number of devices in that grid, which reflects the usage activity of the devices and is an important basis for measuring the "functional activity" of the grid.

[0032] The real-time density of targets in each to-be-simulated grid refers to the density of the number of targets per unit area in each to-be-simulated grid. Here, "target" specifically refers to personnel. This indicator is used to reflect the population distribution and is an important parameter for the digital twin model to perform personnel behavior modeling, flow simulation, and safety warning.

[0033] The real-time average moving speed refers to the average moving speed of all personnel in a certain grid or multiple grids within a time window, which characterizes the activity intensity and flow trend of personnel in the area.

[0034] The preset grid side length refers to the size of the basic grid used to divide the industrial area or environment in the digital twin model. During the simulation process, the industrial area is divided into multiple square grids, and each grid represents an independent computing unit. The setting of the grid side length directly affects the simulation accuracy and computing efficiency, depending on the scale of the target area, the personnel density within the area, the equipment distribution, the mobility, and the real-time data update frequency of the system. Generally, for a relatively small industrial area, the grid side length is set between 40 meters and 100 meters. In this embodiment, it is set to 50 meters to balance the computing accuracy and efficiency, be able to better adapt to the personnel distribution and equipment operation conditions within the area, and provide efficient real-time dynamic optimization within the range allowed by the computing power.

[0035] The preset determination duration refers to the time window used to judge whether the equipment operation rate and personnel density within the grid fluctuate within a certain period of time, depending on the timing characteristics of personnel activities, the operation cycle of the equipment, and the change speed. It is usually set between 5 minutes and 30 minutes. In this embodiment, it is set to 10 minutes, which can effectively detect the fluctuations in personnel flow or equipment operation rate within a short time, facilitating quick response and adjustment.

[0036] The preset correction duration refers to the time window used to correct the optimization model, depending on the length of the optimization cycle, the rate of environmental change, and the convergence time of the model. It is usually set between 30 minutes and 2 hours. In this embodiment, it is set to 30 minutes, which is sufficient to cover the periodic fluctuations of personnel activities and can ensure the timeliness and accuracy of model adjustment, thereby effectively optimizing the grid side length and operation rate fluctuation threshold.

[0037] The simulation optimization index is directly calculated by presetting multiple parameters in the digital twin model. First, the model inputs parameters such as the real-time operation rate, real-time density, and average moving speed of the equipment to simulate the operation state of the entire system. Then, the model calculates the average value of the optimization state values of all grids, and finally obtains a comprehensive optimization index.

[0038] By collecting data such as the real-time operation rate, real-time density, and real-time average moving speed of devices in each grid to be simulated within an industrial area constructed based on a preset grid side length, the dynamic behavior within the industrial area is optimized and published in real time. First, by comparing the real-time operation rate with a preset fluctuation threshold, a number of temporary grids are determined. Then, by analyzing the real-time density of adjacent grids within the temporary grids, the simulation grids are determined, and the prediction optimization index is calculated. The digital twin model is used to simulate the grids to generate the simulation optimization index. Combining the prediction optimization index, the simulation optimization index, and the real-time average moving speed, the grid side length is adjusted to form an optimization model. Finally, the real-time operation rate, real-time density, and real-time average moving speed of each simulation grid are used as input features, and combined with the actual operation situation, the digital twin model is trained to enable it to more accurately predict and simulate the dynamic changes in actual operations. Through this training, the parameters and rules of the model are optimized so that it can better reflect the actual behavior of the target area in different situations, thereby forming an optimized twin model. Finally, based on the optimized twin model, the operation rate fluctuation threshold is adjusted to publish the final optimization model.

[0039] Through dynamic collection and analysis of the real-time density of personnel, the real-time average moving speed of personnel, and the associated device operation rate in each grid within the industrial area, high-precision and adaptive optimization of the digital twin model is achieved. Personnel density reflects the degree of job concentration in a certain area, and the average moving speed reveals the strength of personnel mobility. Together, they constitute the spatial and temporal distribution characteristics of personnel behavior. As an external influencing factor of personnel behavior, the fluctuation of the device operation rate reflects the change in the state of personnel operating the device. Combining density and speed changes can comprehensively determine the work intensity and rhythm of personnel in the area. The "change synchronization degree" is introduced as a natural feedback adjustment mechanism. By measuring the consistency between personnel mobility and regional adjustment behavior, it guides the dynamic update of the grid structure, thereby enabling the twin model to have higher stability and response efficiency in complex environments, significantly improving the intelligent level of personnel behavior simulation, efficiency evaluation, and spatial resource allocation, and effectively solving the problems of slow response speed and low processing accuracy in dealing with dynamic changes due to over-reliance on historical data and single optimization strategies.

[0040] Please continue to refer to Figure 2 as shown, which is the decision logic diagram for determining the temporary grid in this embodiment; The process of determining a number of temporary grids based on the real-time operation rate and the preset operation rate fluctuation threshold within a preset determination duration includes: Calculating the standard deviation of the real-time operation rate to form an operation rate fluctuation value; When the operation rate fluctuation value is greater than the preset operation rate fluctuation threshold, determining the grid to be simulated as a temporary grid to form a number of temporary grids.

[0041] The preset operation rate fluctuation threshold is a standard value used to judge the stability of the operation state of a device or system, depending on the device type, workload, and expected stability requirements in the target area. Generally, the operation rate fluctuation threshold is set between 5% and 20%, and in this embodiment, it is set to 10%. This can ensure the stability of the system while ensuring that the optimization and adjustment mechanism can be started in time when the operation rate changes greatly, improving the response speed and efficiency.

[0042] First, collect the real-time operation rate data of the grid to be simulated within a preset time period, and calculate its standard deviation to obtain the fluctuation value of the operation rate. Then, by comparing the fluctuation value of the real-time operation rate with the preset operation rate fluctuation threshold, if the fluctuation value exceeds the threshold, the grid is determined as a "temporary grid". Through this process, grids with large fluctuations in operation conditions within the preset time period can be identified and used as areas that need further attention and optimization.

[0043] The calculation of the real-time operation rate and the fluctuation value directly reflects the operation stability of the device or area. Therefore, the identification of temporary grids when the fluctuation value exceeds the threshold helps to discover potential operation problems or abnormal areas, enhances the dynamic response ability of the model, can quickly adapt to the changing operation state, and perform targeted optimization and adjustment to improve the overall stability and efficiency of the system.

[0044] Specifically, the process of screening out a number of simulation grids according to the real-time density within any two adjacent temporary grids within the preset determination time period includes: Calculate the standard deviation of the real-time density of each temporary grid to form a density fluctuation value; Screen out a number of simulation grids according to the density fluctuation values of any two adjacent temporary grids.

[0045] Within the preset determination time period, analyze the real-time density within the temporary grids, calculate the standard deviation of the real-time density within each temporary grid to form a density fluctuation value. Then, according to the density fluctuation values of any two adjacent temporary grids, by comparing these fluctuation values, judge which temporary grids can be merged into simulation grids. Specifically, temporary grids with relatively consistent density fluctuation values are determined to be the same simulation grid, which can improve the efficiency and accuracy of the simulation and ensure that the density change within each simulation grid is within a controllable range.

[0046] By comparing the density fluctuation values of adjacent temporary grids, the division of dynamic regions is achieved, effectively improving the rationality and accuracy of the simulation grids. The relationship between the density fluctuation values and the simulation grids reflects the stability and consistency of the personnel distribution. When the density fluctuation is large, it means that there are strong dynamic changes in the region and finer division is required; when the fluctuation is small, it indicates that the distribution of personnel or objects in the region is relatively uniform and can be combined into a larger simulation grid. Precise grid division can be carried out according to real-time changes, which helps to optimize the model simulation effect, improve the calculation efficiency, and ensure the reliability of the simulation results.

[0047] Please continue to refer to Figure 3 as shown, which is the decision logic diagram for determining the simulation grids in this embodiment; The process of screening out a number of simulation grids based on the density fluctuation values of any two adjacent temporary grids includes: Calculating the relative deviation of the density fluctuation values of any two adjacent temporary grids to form a fluctuation consistency; When the fluctuation consistency is greater than a preset consistency threshold, it is determined that the temporary grids are all the simulation grids, forming a number of simulation grids.

[0048] First, calculate the relative deviation of the density fluctuation values of any two adjacent temporary grids with the closest horizontal distance to form a fluctuation consistency. The calculation method of the relative deviation is to compare the difference between the density fluctuation values of the two with their average value to obtain an index measuring the fluctuation consistency. When the fluctuation consistency is greater than the preset consistency threshold, it indicates that these two adjacent temporary grids have a high consistency in the fluctuation of the personnel density. In this way, multiple temporary grids with similar density fluctuation characteristics are determined as simulation grids, and finally a number of simulation grids are formed.

[0049] The preset consistency threshold is the standard for determining whether adjacent temporary grids should be combined into simulation grids. It depends on the requirements for the grid merging accuracy and the characteristics of the target region, such as the law of personnel flow and the amplitude of density change. It is usually set between 0.1 and 0.3. In this embodiment, it is set to 0.2, which can ensure grid merging while avoiding over-simplification, guarantee the rationality and accuracy of the simulation grid division, and ensure that personnel flow simulation can be carried out quickly and accurately in actual operation.

[0050] Judging adjacent temporary grids by the fluctuation consistency ensures the rationality and accuracy of the simulation grids. The relationship between the fluctuation consistency and the density fluctuation value can reflect the stability of the personnel distribution in the area. When the density fluctuations of two adjacent grids are very similar, it indicates that the dynamic changes in these two grids are almost the same. Merging them into one simulation grid can simplify the model calculation without affecting the accuracy. The division of the simulation grids can be dynamically adjusted according to the actual density fluctuation characteristics to optimize the simulation efficiency and accuracy, and avoid unnecessary over-division, thereby improving the overall simulation performance.

[0051] Specifically, the process of calculating the prediction optimization index according to the real-time operation rate and the real-time density in each of the simulation grids includes: Performing a normalization calculation on the real-time density of all the simulation grids to form a normalized density; Performing a normalization calculation on the real-time operation rate of all the simulation grids to form a normalized operation rate; Performing a weighted sum according to a preset density weight, a preset operation rate weight, the normalized density, and the normalized operation rate to form a prediction optimization index.

[0052] The preset density weight refers to the weight value used to adjust the contribution degree of the real-time density to the final optimization result when calculating the prediction optimization index, which depends on the importance of the density in the simulation grid and its impact on system optimization. It is generally set between 0.3 and 0.6. In this embodiment, it is set to 0.4 to ensure the reasonable influence of the density factor in the optimization calculation.

[0053] The preset operation rate weight refers to the weight value used to adjust the contribution degree of the real-time operation rate to the final optimization result when calculating the prediction optimization index. The operation rate reflects the working state and efficiency of the device or system within a certain period of time. The setting of its weight usually depends on the influence degree of the operation efficiency of the device or system on the overall performance. Usually, the device operation rate has a greater impact on system optimization, especially in a large-scale system. Therefore, the operation rate weight is generally set between 0.4 and 0.7. In this embodiment, the operation rate weight is set to 0.6, which can ensure that the operation state of the device occupies an important position in the optimization process, thereby better improving the overall efficiency.

[0054] According to the real-time operation rate and the real-time density of each simulation grid, first perform normalization processing on the real-time density and the real-time operation rate of all simulation grids respectively, and standardize them to a unified numerical range. Then, use the preset density weight and the preset operation rate weight to perform a weighted sum on the normalized real-time density and operation rate, and finally obtain the prediction optimization index. This process ensures that the optimization index can comprehensively reflect the operation efficiency and personnel distribution in the target area by weighted integration of various important factors.

[0055] By comprehensively considering the impacts of real-time density and operation rate factors on the optimization results, the prediction accuracy and operation feasibility can be improved. Through normalization and weighting processes, the impacts of various parameters are quantified and standardized, making the optimization index more in line with actual requirements, effectively evaluating the performance of different simulation grids, and thus providing a reliable basis for subsequent grid adjustment and optimization decisions.

[0056] Specifically, the process of using a preset digital twin model to simulate all the grids to be simulated and generate a simulation optimization index includes: Obtain the real-time operation rate, real-time density, and real-time average moving speed of all the grids to be simulated, and input them into the preset digital twin model to form a simulation optimization index.

[0057] Each grid to be simulated has three key data at a certain moment: real-time operation rate R, real-time density D, and real-time average moving speed. In the preset digital twin model, there is a calculation formula: Q = a×R + b×(1 - D / Dmax) + c×V / Vmax; Q is the single optimization index obtained for a single grid to be simulated. a, b, and c are all pre-determined model weight coefficients, a = 0.5, b = 0.3, c = 0.2. Dmax is the preset maximum value of density, and Vmax is the preset maximum value of average moving speed, which are used for normalization processing.

[0058] Calculate the average value of all the single optimization indexes to obtain the simulation optimization index.

[0059] First, obtain data such as the real-time operation rate, real-time density, and real-time average moving speed of all the grids to be simulated. These data respectively reflect the operation status of the equipment, the personnel density, and the personnel flow situation. Then, input these real-time data into the preset digital twin model. The model performs simulation calculations based on these input information and finally generates a simulation optimization index. This index comprehensively reflects the current operation efficiency and personnel flow status of the system, thus providing data support for further optimization.

[0060] Introducing the real-time average moving speed as an additional input parameter can significantly enhance the accuracy and dynamic adaptability of the digital twin model in the simulation and optimization processes. The real-time operation rate and real-time density mainly reflect the load situation of the equipment and the distribution density of personnel, but these parameters fail to fully capture the changes and dynamic factors of personnel flow. And the average moving speed of personnel, as a dynamic factor, can provide information about the flow trend and behavior pattern of personnel in space, which is crucial for analyzing the migration and distribution of personnel between different grids.

[0061] When the real-time average moving speed of personnel is incorporated into the model, the system can more accurately reflect the dynamic behavior of personnel, thereby optimizing the matching between personnel distribution and equipment load, and avoiding congestion or uneven load problems that may occur by simply relying on static density and equipment operation rate. The mobility of personnel can affect the load of a certain grid. Especially in high-mobility areas, the degree of aggregation of personnel at different times and the load of equipment may change drastically. By comprehensively considering the real-time moving speed, the model can not only capture the distribution of personnel within the grid, but also track the flow state of personnel between different grids, making the optimization process more in line with the dynamic changes of the actual environment. Therefore, adding the factor of real-time average moving speed enables the model to adaptively adjust the grid boundary and operation strategy when facing a complex and dynamic working environment, thereby achieving more refined and efficient optimization. This not only improves the accuracy of the overall optimization index, but also effectively responds to various changes in the actual scenario, enhancing the reliability and efficiency of system operation.

[0062] By introducing the real-time average moving speed as the third parameter, the model can more comprehensively reflect the dynamic behavior of personnel, that is, the moving speed of personnel. This additional dynamic parameter enables the model not to stay at the analysis of static data, but to further optimize the system simulation by combining the flow of personnel. Specifically, the introduction of the moving speed enables the model to identify the flow trend of personnel between different grids, further optimizing the matching between personnel density and equipment operation efficiency, and avoiding congestion or inefficiency problems caused by excessive personnel density or heavy equipment load.

[0063] Specifically, the process of adjusting the preset grid side length according to the predicted optimization index and the simulated optimization index includes: Calculating the relative deviation between the predicted optimization index and the simulated optimization index to form an index deviation; Adjusting the preset grid side length according to the index deviation and the real-time average moving speed.

[0064] In the process of adjusting the preset grid side length, first calculate the relative deviation between the predicted optimization index and the simulated optimization index to obtain the index deviation, and then adjust the preset grid side length based on this index deviation and the real-time average moving speed. By comparing the differences between the predicted results and the simulated results and combining the flow speed of personnel, the size of the grid is optimized in real time, thereby improving the response ability and accuracy of the overall system.

[0065] By introducing the comparison between the prediction optimization index and the simulation optimization index, the deviation of the current grid design in practical applications can be effectively evaluated. The index deviation reflects the difference between the prediction result and the actual simulation result, which is the basis for adjusting the grid side length, provides the gap between the model prediction and the actual situation, helps to identify the optimization space, and by combining the real-time average moving speed, the impact of personnel mobility is considered, helping to adjust the grid side length to adapt to the personnel dynamics in actual operation.

[0066] Specifically, the process of adjusting the preset grid side length according to the index deviation and the real-time average moving speed includes: When the deviation comparison result shows that the index deviation is greater than the preset deviation threshold, increase the preset grid side length according to the relative deviation between the index deviation and the preset deviation threshold and the preset adjustment coefficient to form a temporary grid side length; Adjust the temporary grid side length according to all the temporary grid side lengths within the preset adjustment duration and the real-time average moving speed to form an adjusted grid side length.

[0067] The preset deviation threshold is a standard for measuring the deviation between the prediction optimization index and the simulation optimization index, which depends on the system's accuracy requirements, historical data, and industry standards. It is usually set between 0.5% and 10%. In this embodiment, it is set to 5%, which can ensure that the adjustment of the grid side length is only carried out when the difference between the prediction optimization index and the simulation optimization index is large without causing frequent adjustments, thereby improving the calculation efficiency and resource utilization rate.

[0068] The preset adjustment coefficient is a weighting factor for adjusting the grid side length when the deviation exceeds the preset threshold, which depends on the system's stability requirements and error tolerance. It is usually set between 0.1 and 1.0. In this embodiment, the preset adjustment coefficient is set to 0.5, which can flexibly adjust the grid side length according to the deviation situation on the premise of ensuring the stable operation of the system, ensuring a more accurate grid division.

[0069] During the process of adjusting the preset grid side length, first judge whether the index deviation is greater than the preset deviation threshold. If so, increase the preset grid side length according to the relative deviation between the index deviation and the preset deviation threshold, and the preset adjustment coefficient, to form a temporary grid side length. Then, within the preset adjustment duration, comprehensively consider all the temporary grid side lengths and the real-time average moving speed, and further adjust the temporary grid side length to finally obtain the adjusted grid side length, ensuring that the grid size is adjusted in real time according to the actual deviation and the dynamic changes of personnel flow.

[0070] By combining the use of exponential deviation and real-time average moving speed, a more refined grid adjustment strategy is achieved. The exponential deviation provides a measure of the difference between the model prediction and the actual situation, making the grid adjustment more targeted. When the exponential deviation is greater than the preset deviation threshold, the grid side length is adjusted to quickly respond to the deviation between the model and reality, thereby improving the model's accuracy and precision. Introducing the real-time average moving speed as a dynamic factor, considering the impact of personnel flow on the grid distribution, can further optimize the grid adjustment to ensure that it can adapt to the actual personnel flow situation, avoid overly large or small grid divisions, and thus improve the efficiency of resource scheduling.

[0071] Please continue to refer to Figure 4 as shown, which is the determination logic diagram for forming the adjusted grid side length in this embodiment; Adjusting the temporary grid side length according to all the temporary grid side lengths and the real-time average moving speed within the preset adjustment duration, the process of forming the adjusted grid side length includes: Calculating the standard deviation of the formation times of all the temporary grid side lengths and normalizing it to form a normalized number of times fluctuation value; Calculating the standard deviation of all the real-time average moving speeds and normalizing it to form a normalized moving speed fluctuation value; Calculating the correlation coefficient between the normalized number of times fluctuation value and the normalized moving speed fluctuation value to form a change synchronization degree; When the change synchronization degree is less than the preset standard synchronization degree, increasing the temporary grid side length according to the relative deviation between the preset standard synchronization degree and the change synchronization degree and the preset adjustment coefficient to form the adjusted grid side length.

[0072] The preset adjustment duration refers to the time range allowed for grid side length adjustment during the adjustment process, which depends on system response requirements, operation stability, and business requirements. It is usually set between 5 minutes and 40 minutes. In this embodiment, it is set to 30 minutes, which can effectively balance the system's response speed and operation stability, ensure that each grid side length adjustment is based on sufficient real-time data analysis, and avoid the unstable impact caused by excessive adjustment.

[0073] The preset standard synchronization degree is a standard value used to measure the degree of synchronous change between the number of changes in the temporary grid side length and the fluctuation of the real-time average moving speed during the grid side length adjustment process. It depends on system stability requirements and adjustment responsiveness and is usually set between 0.5 and 0.9. In this embodiment, it is set to 0.8, which can balance the system's response ability to minor changes and the system's stability.

[0074] First, calculate the standard deviation of the formation times of all temporary grid side lengths and normalize it to obtain the normalized frequency fluctuation value. Also, calculate the standard deviation of all real-time average moving speeds and normalize it to obtain the normalized speed fluctuation value. Then, calculate the correlation coefficient between these two fluctuation values to form the change synchronization degree, which reflects the synchronization relationship between the change in grid side length and the moving speed of personnel. When the change synchronization degree is lower than the preset standard synchronization degree, increase the temporary grid side length according to the relative deviation between the preset standard synchronization degree and the change synchronization degree, as well as the preset adjustment coefficient, and finally form the adjusted grid side length.

[0075] By normalizing the formation times of the temporary grid side lengths and the fluctuations of the real-time average moving speeds, and measuring their change synchronization degree, the relationship between the grid side length and the personnel flow can be captured more accurately. This process ensures the adaptability of the model in the face of dynamic changes and avoids incorrect predictions caused by excessive fluctuations or out-of-synchronization. The normalized frequency fluctuation value reflects the change in the grid side length and shows the impact of system state or environmental changes on grid division. The normalized speed fluctuation value reflects the fluctuations in the dynamic moving speed of personnel and reveals the impact of personnel mobility on the entire area. The change synchronization degree provides the correlation between the two and measures the matching degree between personnel mobility and grid division, ensuring that in practical applications, grid adjustment and personnel dynamics can change synchronously.

[0076] Specifically, the process of correcting the preset operation rate fluctuation threshold according to the simulation optimization index re-simulated based on the optimized twin model within the preset correction duration includes: Calculate the standard deviation of the simulation optimization index to form the optimization index fluctuation value; When the optimization index fluctuation value is greater than the preset index fluctuation threshold, reduce the preset operation rate fluctuation threshold according to the relative deviation between the optimization index fluctuation value and the preset index fluctuation threshold, as well as the preset correction coefficient, to form the corrected operation rate fluctuation threshold.

[0077] The preset index fluctuation threshold refers to the maximum range of allowable fluctuations of the simulation optimization index in the optimized twin model, which depends on the requirements of the actual application scenario and is mainly based on the fluctuation range tolerated by the system and the stability of historical data. It is usually set between 0.1 and 0.2, and in this embodiment, it is set to 0.15, which can ensure that the model avoids unnecessary frequent adjustments during the correction process while ensuring the stability of the system.

[0078] The preset correction coefficient refers to the proportional coefficient for adjusting the preset operation rate fluctuation threshold according to the deviation of the optimization index fluctuation value when correcting the operation rate fluctuation threshold, which reflects the sensitivity of the optimization model to correct fluctuations and depends on the response speed and accuracy requirements of the system for fluctuation adjustment. Usually, the setting is between 0.2 and 0.3. In this embodiment, it is set to 0.25, which can make the correction of the fluctuation value more stable and avoid the instability of the model caused by large fluctuations.

[0079] By re-simulating the optimization twin model, a new simulated optimization index is obtained, and the standard deviation of this optimization index is calculated to form the optimization index fluctuation value. Next, this fluctuation value is compared with the preset index fluctuation threshold. If the optimization index fluctuation value is greater than the preset threshold, the preset operation rate fluctuation threshold is adjusted according to their relative deviation, so as to obtain the corrected operation rate fluctuation threshold. Through dynamic adjustment, the adaptability of the model to the actual operation conditions is ensured, and the optimization effect is improved.

[0080] By calculating the fluctuation value of the simulated optimization index and comparing it with the preset threshold, the operation rate fluctuation threshold can be dynamically adjusted according to the deviation, ensuring that the model can more accurately adapt to the changes in the actual operation environment. The self-correction mechanism improves the flexibility and adaptability of the system, avoids errors caused by overly strict or loose threshold settings, ensures the efficient and accurate operation of the model in different scenarios, and thus improves the overall optimization ability and stability.

[0081] So far, the technical solution of the present invention has been described in conjunction with the preferred embodiments shown in the drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A dynamic optimization and release method for digital twin models, characterized in that, Including: Collecting the real-time operation rate of several devices in each to-be-simulated grid within an industrial area constructed based on a preset grid side length, the real-time density of targets in each to-be-simulated grid, and the real-time average moving speed; Determining several temporary grids according to the real-time operation rate within a preset determination duration and a preset operation rate fluctuation threshold; Screening out several simulation grids according to the real-time density within any two adjacent temporary grids within a preset determination duration; Calculating a prediction optimization index according to the real-time operation rate and the real-time density within each simulation grid; Using a preset digital twin model to simulate all the to-be-simulated grids and generating a simulation optimization index; Adjusting the preset grid side length according to the prediction optimization index and the simulation optimization index to form an adjusted grid side length; Training the preset digital twin model using the real-time operation rate, the real-time density, and the real-time average moving speed of all the simulation grids determined based on the adjusted grid side length to form an optimized twin model; Correcting the preset operation rate fluctuation threshold according to the simulation optimization index re-simulated based on the optimized twin model within a preset correction duration to form a corrected operation rate fluctuation threshold; Releasing the optimized twin model re-formed based on the corrected operation rate fluctuation threshold.

2. The dynamic optimization and publishing method for the digital twin model according to claim 1, characterized in that, The process of determining several temporary grids according to the real-time operation rate within a preset determination duration and a preset operation rate fluctuation threshold includes: Calculating the standard deviation of the real-time operation rate to form an operation rate fluctuation value; When the operation rate fluctuation value is greater than the preset operation rate fluctuation threshold, determining the to-be-simulated grid as a temporary grid to form several temporary grids.

3. The dynamic optimization and publishing method for the digital twin model according to claim 2, characterized in that, The process of screening out several simulation grids according to the real-time density within any two adjacent temporary grids within a preset determination duration includes: Calculating the standard deviation of the real-time density of each temporary grid to form a density fluctuation value; Screening out several simulation grids according to the density fluctuation values of any two adjacent temporary grids.

4. The dynamic optimization and release method for the digital twin model according to claim 3, characterized in that The process of screening out several simulation grids according to the density fluctuation values of any two adjacent temporary grids includes: Calculating the relative deviation of the density fluctuation values of any two adjacent temporary grids to form a fluctuation consistency; When the fluctuation consistency is greater than a preset consistency threshold, determining that the temporary grids are all the simulation grids to form several simulation grids.

5. The dynamic optimization and publishing method for the digital twin model according to claim 4, wherein The process of calculating a prediction optimization index according to the real-time operation rate and the real-time density within each simulation grid includes: Performing normalization calculation on the real-time density of all the simulation grids to form a normalized density; Performing normalization calculation on the real-time operation rate of all the simulation grids to form a normalized operation rate; Performing weighted summation according to a preset density weight, a preset air flow rate weight, the normalized density, and the normalized operation rate to form a prediction optimization index.

6. The dynamic optimization and publishing method for the digital twin model according to claim 5, characterized in that, The process of using a preset digital twin model to simulate all the to-be-simulated grids and generating a simulation optimization index includes: Obtaining the real-time operation rate and the real-time density of all the to-be-simulated grids and inputting them into the preset digital twin model to form a simulation optimization index.

7. The dynamic optimization and publishing method for the digital twin model according to claim 6, characterized in that, The process of adjusting the preset grid side length according to the predicted optimization index and the simulated optimization index includes: Calculating the relative deviation between the predicted optimization index and the simulated optimization index to form an index deviation; Adjusting the preset grid side length according to the index deviation and the real-time average moving speed.

8. The dynamic optimization and publishing method for the digital twin model according to claim 7, wherein The process of adjusting the preset grid side length according to the index deviation and the real-time average moving speed includes: When the deviation comparison result shows that the index deviation is greater than the preset deviation threshold, increasing the preset grid side length according to the relative deviation between the index deviation and the preset deviation threshold and a preset adjustment coefficient to form a temporary grid side length; Adjusting the temporary grid side length according to all the temporary grid side lengths within a preset adjustment duration and the real-time average moving speed to form an adjusted grid side length.

9. The dynamic optimization and publishing method for the digital twin model according to claim 8, wherein The process of adjusting the temporary grid side length according to all the temporary grid side lengths within a preset adjustment duration and the real-time average moving speed to form an adjusted grid side length includes: Calculating the standard deviation of the formation times of all the temporary grid side lengths and normalizing it to form a normalized number fluctuation value; Calculating the standard deviation of all the real-time average moving speeds and normalizing it to form a normalized speed fluctuation value; Calculating the correlation coefficient between the normalized number fluctuation value and the normalized speed fluctuation value to form a change synchronization degree; When the change synchronization degree is less than the preset standard synchronization degree, increasing the temporary grid side length according to the relative deviation between the preset standard synchronization degree and the change synchronization degree and the preset adjustment coefficient to form an adjusted grid side length.

10. The dynamic optimization and release method for the digital twin model according to claim 9, characterized in that, The process of correcting the preset operation rate fluctuation threshold according to the simulated optimization index re-simulated based on the optimized twin model within a preset correction duration to form a corrected operation rate fluctuation threshold includes: Calculating the standard deviation of the simulated optimization index to form an optimization index fluctuation value; When the optimization index fluctuation value is greater than the preset index fluctuation threshold, decreasing the preset operation rate fluctuation threshold according to the relative deviation between the optimization index fluctuation value and the preset index fluctuation threshold and a preset correction coefficient to form a corrected operation rate fluctuation threshold.

Images