Method and system for monitoring and optimizing the status of three-dimensional smart parks

Abstract

A method and system for monitoring and optimizing the status of a three-dimensional smart park include: acquiring park image videos; utilizing a management prediction model to extract unit areas in the park video image and assigning a dynamic identifier to each unit area, and constructing a time series of the dynamic identifiers of each unit area according to the time sequence of each frame; judging the real-time status of the unit area based on the state change of the unit area in the current frame sequence; utilizing a management prediction model to analyze the state change trend of the unit area; generating optimization suggestions based on the analysis results, associating the optimization suggestions with the dynamic identifiers of the unit area to generate an optimization data sequence; mapping the dynamic identifiers, real-time status, and optimization suggestions of the unit area to the corresponding unit area in the three-dimensional real-life model; and highlighting abnormal areas in the three-dimensional model in the form of dynamic identification in combination with the dynamically updated unit area status data, and synchronously updating the remote notification content to trigger the corresponding emergency management mechanism.

Description

Technical Field

[0001] The present invention belongs to the technical field of new energy status monitoring, and more specifically, relates to a method and system for three-dimensional smart park status monitoring and status optimization. Background Art

[0002] New energy status monitoring is a critical component of smart parks, providing reliable data support for management and decision-making within them, enabling timely responses to emergencies in new energy facilities and other facilities within the park. Currently, smart park management systems rely solely on cameras for video surveillance of facilities, roads, and other areas within the park. Management and decision-making are performed manually based on incomplete camera footage. This can lead to discrepancies between actual conditions and video footage, as well as synchronization issues between monitored and displayed images, resulting in poor park management effectiveness.

[0003] In the existing technology, by combining three-dimensional visualization presentation capabilities, it only assists the park to improve the management system of new energy facilities, realizes effective management models and adjustment mechanisms for energy conservation and consumption reduction, low carbon emissions, etc. of the park, and improves the energy efficiency and energy supply service level of the park. However, only managing new energy facilities ignores the impact of other facilities in the park on energy conservation and consumption reduction, low carbon emissions, etc. of the park, and also ignores the impact of various wastes generated on the overall function of the park. It is still impossible to achieve comprehensive intelligent management of the park and improve the comprehensive energy efficiency of the park. Summary of the Invention

[0004] In order to solve the deficiencies in the prior art, the present invention provides a three-dimensional smart park management method and system.

[0005] The present invention adopts the following technical solutions.

[0006] A first aspect of the present invention provides a method for monitoring and optimizing the status of a three-dimensional smart park. The park's status monitoring objects include parking spaces, equipment, and waste storage areas. The equipment includes new energy equipment and traditional energy equipment. The method includes:

[0007] Step 1, obtain the park image video; establish a management prediction model based on the BP neural network; use the trained management prediction model to extract the unit area from the park image video and assign a dynamic identifier to each unit area, the dynamic identifier includes: the coordinates of the center point of the unit area, the coordinates of the boundary of the unit area and the state change of the unit area, wherein the state change of the unit area includes the parking state quantity, the equipment state quantity, and the waste state quantity; when the equipment in the unit area is operating normally, the equipment state quantity is set to 1, otherwise the equipment state quantity is set to 0; when the equipment state quantity is 1 and the parking spaces in the unit area are occupied by 70% or more, or the equipment state quantity is 0 and the parking spaces in the unit area are occupied by 50% or more, the parking state quantity is set to 1, otherwise the parking state quantity is set to 0; when the equipment state quantity is 1 and the filling quantity of the waste storage area in the unit area is 70% or more, or the equipment state quantity is 0 and the filling quantity of the waste storage area in the unit area is 50% or more, the waste state quantity is set to 1, otherwise the waste state quantity is set to 0;

[0008] Based on the dynamic mapping method, the unit area and the dynamic identifier of each unit area corresponding to each frame in the frame sequence of the park image video are determined, and the time sequence of the dynamic identifier of each unit area is constructed according to the time sequence of each frame;

[0009] Step 2: Extract the dynamic identifier of the target unit area during the monitoring period from the time series of the dynamic identifiers of each unit area; if the parking state variable of the extracted dynamic identifier is 1, generate a vehicle dispatch request signal for the target unit area; if the equipment state variable is 1, generate an equipment inspection request signal for the target unit area; if the waste state variable is 1, generate a waste removal request signal for the target unit area; if the parking state variable, equipment state variable, and waste state variable are all 1, generate an alarm signal for the target unit area;

[0010] In step 3, the host computer generates corresponding management measures according to the received request signal, and uses the trained management prediction model to output the dynamic identifiers of the target unit area and the N unit areas adjacent to the target unit area after the implementation of the management measures; the park status is optimized by the management measures corresponding to the minimum sum of the weighted sum of the parking status quantity, equipment status quantity and waste status quantity of each unit area.

[0011] Preferably, step 1 comprises:

[0012] Fixed cameras and drone-based dynamic data acquisition equipment deployed throughout the park capture real-time video data covering the entire park. Wireless communication units are used to transmit the video data to the main control terminal, which performs pre-processing operations on the video data, including noise removal, frame rate standardization, and color enhancement.

[0013] Based on parking space occupancy records, equipment operation logs, and historical management data from waste storage areas, core features were extracted and labeled to construct a data set for training a management prediction model using a BP neural network. The labeled data set was trained using the BP neural network to generate a management prediction model. The classification capability of the management prediction model was optimized using a random forest classification model, and the weight parameters of the management prediction model were adjusted through cross-validation.

[0014] The image and video data are analyzed frame by frame, and the trained management prediction model is applied to identify the unit areas in the park. The unit areas include parking spaces, equipment areas, and waste storage areas. For each identified unit area, a dynamic identifier corresponding to the dynamic state is generated. The dynamic identifier includes the coordinates of the center point of the unit area, the coordinates of the boundary of the unit area, and the state change of the unit area.

[0015] Preferably, the dynamic identifier includes geometric information and dynamic status information of the unit area, wherein the geometric information includes the center point coordinates and boundary coordinates of the unit area, and the dynamic status information includes the real-time occupancy status of the parking space, the operating status of the equipment, and the filling status of the waste storage area;

[0016] Dynamic identifiers are generated by analyzing the morphological features and spatial distribution of unit areas in video frames. Parking spaces are identified by regularly arranged rectangular areas, equipment areas by morphological features with regular boundaries, and waste storage areas by dynamically changing boundary contours and fill ratios.

[0017] The dynamic identifier of each unit area is dynamically mapped to the video frame sequence, and the position change, boundary adjustment and state change information of each unit area in the video are recorded.

[0018] Preferably, the step of dynamically mapping the dynamic identifier of each unit area with the video frame sequence and recording the position change, boundary adjustment, and state change information of each unit area in the video includes:

[0019] In continuous video frames, the unit area is tracked across frames according to dynamic mapping rules. Position continuity rules are set to limit the offset of the center point coordinates to no more than five pixels. Boundary morphological similarity rules are set to require the boundary overlap rate to reach more than 70 percent. The dynamic identifiers of the unit area are combined to match the status data of parking spaces, equipment areas, and waste storage areas.

[0020] The geometric change information of the unit area is recorded through cross-frame tracking, including the dynamic adjustment of the center point coordinates and the shape change of the boundary contour; the state change amount of the unit area is recorded. When the equipment in the unit area is operating normally, the equipment state amount is set to 1; when the equipment stops abnormally, the equipment state amount is set to 0; when the equipment state amount is 1 and the occupancy ratio of parking spaces in the unit area reaches 70% or above, or the equipment state amount is 0 and the occupancy ratio of parking spaces reaches 50% or above, the parking state amount is set to 1, otherwise it is set to 0; when the equipment state amount is 1 and the filling amount of the waste storage area in the unit area reaches 70% or above, or the equipment state amount is 0 and the filling amount of the waste storage area reaches 50% or above, the waste state amount is set to 1, otherwise it is set to 0; the state change amount of the unit area is used to dynamically reflect the changes in the parking, equipment and waste storage states of the unit area;

[0021] Position changes, boundary outline updates, and dynamic state transitions are integrated into a time series along the time axis.

[0022] Preferably, step 2 includes:

[0023] Analyze the state change trend of the target unit area based on the unit area dynamic identifier time series;

[0024] Generate a real-time status record based on the dynamic changes of the parking state quantity, the equipment state quantity and the waste state quantity, wherein the real-time status record includes a state change curve and anomaly detection results of the target unit area during the monitoring period;

[0025] Anomaly detection results are generated by analyzing the behavior of exceeding the limit of parking space occupancy time, the deviation of equipment operating parameters from the normal range, and the continuous growth trend of the filling rate of the waste storage area;

[0026] The real-time status records are bound to the time series of dynamic identifiers through dynamic mapping method.

[0027] Preferably, the real-time status record includes a dynamic identifier of the target unit area and a state change amount; a management request signal is generated according to the state change amount, and the management request signal includes:

[0028] When the parking status change trend shows that the occupancy rate continues to exceed the set threshold, a vehicle dispatch request signal is generated. The signal content includes a dynamic identifier, a parking space occupancy rate change trend, and a vehicle dispatch suggestion.

[0029] When the change trend of the equipment status quantity shows that the frequency of the operating parameters exceeds the preset range, an equipment inspection request signal is generated. The signal content includes the dynamic identifier, equipment operating parameters and fault prediction results;

[0030] When the change trend of the waste state quantity shows that the filling rate growth rate exceeds the set value, a waste cleaning request signal is generated, and the signal content includes a dynamic identifier, the filling growth trend of the waste storage area and a cleaning time suggestion;

[0031] The above management request signal is sent to the management terminal through the remote communication module, triggering the corresponding management measures of the park and synchronously recorded as optimization data records.

[0032] Preferably, step 3 includes:

[0033] After receiving the request signal from the target unit area, the host computer extracts the real-time status data and historical time series data of the target unit area and its N adjacent unit areas through the management prediction model, and uses multi-objective optimization to calculate the change trends of the parking state quantity, equipment state quantity, and waste state quantity respectively;

[0034] Based on the trend calculation results, a dynamic weighting strategy is used to adjust the state quantity weight. The parking state quantity weight is adjusted according to the priority of vehicle scheduling needs. The equipment state quantity weight changes dynamically according to the equipment operation abnormality rate. The waste state quantity weight is calculated based on the filling growth rate.

[0035] During the optimization calculation process, the state quantities of adjacent unit areas and the target unit area are analyzed collaboratively, giving priority to the feasibility of vehicle scheduling due to parking space saturation, the coverage of equipment inspections, and the time and resource constraints of waste removal paths.

[0036] Generate a resource allocation plan for the target unit area and its adjacent areas through a state quantity weighted sum minimization optimization model. The resource allocation plan includes vehicle scheduling target area allocation, equipment inspection route optimization, and waste cleanup time scheduling.

[0037] The resource allocation plan is bound to the dynamic identifier of the unit area, the state change in the dynamic identifier is updated, and an optimized dynamic identifier sequence is generated to optimize the park state.

[0038] Preferably, the generation of the resource allocation scheme includes:

[0039] For parking resource allocation, the shortest path method is used to plan vehicle dispatch routes based on the dynamic identifiers of the target unit area and its adjacent areas, combined with vehicle flow trends and real-time parking space occupancy rates calculated from historical monitoring data. Path optimization takes into account the distance from the vehicle's current location to the target parking area, the parking space availability rate, and the priority of parking needs.

[0040] For resource allocation of equipment inspections, we utilize time series data of equipment operating parameters, combined with equipment operating anomaly rates, to generate an inspection plan through an inspection path optimization algorithm. This inspection plan prioritizes the degree of equipment anomalies and the path with the shortest inspection time.

[0041] For waste disposal resource allocation, the priority and arrival order of disposal vehicles are assigned based on the waste storage area's fill rate growth trend and the disposal time window. The disposal route optimization is based on the waste storage area's full load time prediction, the disposal vehicle's operating capacity, and the accessibility of the logistics channel.

[0042] The resource allocation scheme updates the state change of the target unit area and its adjacent areas through dynamic identifiers, generates an optimized dynamic identifier sequence and applies it.

[0043] Preferably, the optimized dynamic identifier sequence and application includes:

[0044] Using an optimized dynamic identifier sequence, the status display of unit areas in the park's 3D model is updated in real time. Parking status is dynamically displayed in color-coded form, equipment status is displayed with boundary highlights and operating parameter curves superimposed, and waste status is dynamically displayed with a fill transparency gradient.

[0045] By linking resource allocation plans with dynamic identifiers of corresponding unit areas in the 3D model through multi-dimensional structured data, the execution plan for vehicle dispatch, equipment inspection, and waste disposal is visualized, allowing managers to intuitively view status changes and optimization measures for each unit area.

[0046] When the state change amount in the dynamic identifier sequence triggers the preset threshold, a new management request signal is generated in combination with the real-time state data to start the next round of resource optimization and state adjustment.

[0047] A second aspect of the present invention provides a three-dimensional smart park status monitoring and optimization system, comprising: a data acquisition and processing module, a status analysis and decision module, and an optimization output module;

[0048] The data acquisition and processing module is used to collect park image videos through cameras and drones, pre-process the video data, generate input data for identifying unit areas, and extract the dynamic identifier of the unit area. The dynamic identifier includes the coordinates of the center point of the unit area, the coordinates of the boundary, and the state change amount;

[0049] The state analysis and decision-making module is used to analyze the state changes of the target unit area and its adjacent areas based on the time series of dynamic identifiers, generate management request signals for vehicle scheduling, equipment inspection and waste disposal, and calculate the optimized dynamic identifiers and corresponding management measures in combination with the management prediction model;

[0050] The optimization output module is used to output the optimized dynamic identifiers and management measures to the management terminal to dynamically display the real-time status, historical changes and optimization suggestions of the unit area, while continuously generating real-time monitoring and management requests for new status changes.

[0051] Compared with the prior art, the beneficial effects of the present invention include at least:

[0052] (1) The present invention includes parking spaces, traditional energy equipment, and waste storage areas in the park into the monitoring scope, achieving comprehensive status monitoring and optimization of various resources in the park, overcoming the limitation of the existing technology that only manages new energy facilities;

[0053] (2) By acquiring park image videos and combining them with the construction of dynamic identifiers and time series analysis, real-time resource status monitoring is achieved. Furthermore, through weighting and optimization of status quantities, management measures for vehicle scheduling, equipment inspection, and waste disposal are dynamically generated, improving the utilization efficiency and response speed of park resources.

[0054] (3) The present invention presents the real-time status and optimization suggestions of the park in a three-dimensional model through the construction and real-time update of dynamic identifiers, assisting managers in making decisions in an intuitive way and solving the problem of the monitoring screen being out of sync with the actual situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a structural diagram of a three-dimensional smart park management method provided in accordance with an embodiment of the present invention;

[0056] Figure 2 It is a structural diagram of a three-dimensional smart park management system provided in accordance with an embodiment of the present invention. DETAILED DESCRIPTION

[0057] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without making creative efforts are all within the scope of protection of the present invention.

[0058] like Figure 1 As shown, embodiment 1 of the present invention provides a method for monitoring and optimizing the status of a three-dimensional smart park. The park's status monitoring objects include parking spaces, equipment, and waste storage areas. The equipment includes new energy equipment and traditional energy equipment; including:

[0059] Acquire park image videos; establish a management prediction model based on a BP neural network; use the trained management prediction model to extract unit areas from the park image videos and assign dynamic identifiers to each unit area, the dynamic identifiers include: the coordinates of the center point of the unit area, the coordinates of the boundary of the unit area, and the state change of the unit area, wherein the state change of the unit area includes the parking state quantity, the equipment state quantity, and the waste state quantity; when the equipment in the unit area is operating normally, the equipment state quantity is set to 1, otherwise the equipment state quantity is set to 0; when the equipment state quantity is 1 and the parking spaces in the unit area are occupied by 70% or more, or the equipment state quantity is 0 and the parking spaces in the unit area are occupied by 50% or more, the parking state quantity is set to 1, otherwise the parking state quantity is set to 0; when the equipment state quantity is 1 and the filling quantity of the waste storage area in the unit area is 70% or more, or the equipment state quantity is 0 and the filling quantity of the waste storage area in the unit area is 50% or more, the waste state quantity is set to 1, otherwise the waste state quantity is set to 0;

[0060] Specifically, the benefit of this setup is that it links parking scheduling and waste removal with equipment operation, optimizing parking management and waste management while ensuring reliable energy supply for the park, improving the park's overall energy efficiency, and achieving the management goals of energy conservation, consumption reduction, and low carbon emissions for the entire park.

[0061] Based on the dynamic mapping method, the unit area and the dynamic identifier of each unit area corresponding to each frame in the frame sequence of the park image video are determined, and the time sequence of the dynamic identifier of each unit area is constructed according to the time sequence of each frame;

[0062] Preferably, step 1 comprises:

[0063] Fixed cameras and drone-based dynamic data acquisition equipment deployed throughout the park capture real-time video data covering the entire park. Wireless communication units are used to transmit the video data to the main control terminal, which performs pre-processing operations on the video data, including noise removal, frame rate standardization, and color enhancement.

[0064] Based on parking space occupancy records, equipment operation logs, and historical management data from waste storage areas, core features were extracted and labeled to construct a data set for training a management prediction model using a BP neural network. The labeled data set was trained using the BP neural network to generate a management prediction model. The classification capability of the management prediction model was optimized using a random forest classification model, and the weight parameters of the management prediction model were adjusted through cross-validation.

[0065] The image and video data are analyzed frame by frame, and the trained management prediction model is applied to identify the unit areas in the park. The unit areas include parking spaces, equipment areas, and waste storage areas. For each identified unit area, a dynamic identifier corresponding to the dynamic state is generated. The dynamic identifier includes the coordinates of the center point of the unit area, the coordinates of the boundary of the unit area, and the state change of the unit area.

[0066] Preferably, the dynamic identifier includes geometric information and dynamic status information of the unit area, wherein the geometric information includes the center point coordinates and boundary coordinates of the unit area, and the dynamic status information includes the real-time occupancy status of the parking space, the operating status of the equipment, and the filling status of the waste storage area;

[0067] Dynamic identifiers are generated by analyzing the morphological features and spatial distribution of unit areas in video frames. Parking spaces are identified by regularly arranged rectangular areas, equipment areas by morphological features with regular boundaries, and waste storage areas by dynamically changing boundary contours and fill ratios.

[0068] The dynamic identifier of each unit area is dynamically mapped to the video frame sequence, and the position change, boundary adjustment and state change information of each unit area in the video are recorded.

[0069] Preferably, the step of dynamically mapping the dynamic identifier of each unit area with the video frame sequence and recording the position change, boundary adjustment, and state change information of each unit area in the video includes:

[0070] In continuous video frames, the unit area is tracked across frames according to dynamic mapping rules. Position continuity rules are set to limit the offset of the center point coordinates to no more than five pixels. Boundary morphological similarity rules are set to require the boundary overlap rate to reach more than 70 percent. The dynamic identifiers of the unit area are combined to match the status data of parking spaces, equipment areas, and waste storage areas.

[0071] The geometric change information of the unit area is recorded through cross-frame tracking, including the dynamic adjustment of the center point coordinates and the shape change of the boundary contour; the state change amount of the unit area is recorded. When the equipment in the unit area is operating normally, the equipment state amount is set to 1; when the equipment stops abnormally, the equipment state amount is set to 0; when the equipment state amount is 1 and the occupancy ratio of parking spaces in the unit area reaches 70% or above, or the equipment state amount is 0 and the occupancy ratio of parking spaces reaches 50% or above, the parking state amount is set to 1, otherwise it is set to 0; when the equipment state amount is 1 and the filling amount of the waste storage area in the unit area reaches 70% or above, or the equipment state amount is 0 and the filling amount of the waste storage area reaches 50% or above, the waste state amount is set to 1, otherwise it is set to 0; the state change amount of the unit area is used to dynamically reflect the changes in the parking, equipment and waste storage states of the unit area;

[0072] Position changes, boundary outline updates, and dynamic state transitions are integrated into a time series along the time axis.

[0073] Step 2: Extract the dynamic identifier of the target unit area during the monitoring period from the time series of the dynamic identifiers of each unit area; if the parking state variable of the extracted dynamic identifier is 1, generate a vehicle dispatch request signal for the target unit area; if the equipment state variable is 1, generate an equipment inspection request signal for the target unit area; if the waste state variable is 1, generate a waste removal request signal for the target unit area; if the parking state variable, equipment state variable, and waste state variable are all 1, generate an alarm signal for the target unit area;

[0074] Preferably, step 2 includes:

[0075] Analyze the state change trend of the target unit area based on the unit area dynamic identifier time series;

[0076] Generate a real-time status record based on the dynamic changes of the parking state quantity, the equipment state quantity and the waste state quantity, wherein the real-time status record includes a state change curve and anomaly detection results of the target unit area during the monitoring period;

[0077] Anomaly detection results are generated by analyzing the behavior of exceeding the limit of parking space occupancy time, the deviation of equipment operating parameters from the normal range, and the continuous growth trend of the filling rate of the waste storage area;

[0078] The real-time status records are bound to the time series of dynamic identifiers through dynamic mapping method.

[0079] Preferably, the real-time status record includes a dynamic identifier of the target unit area and a state change amount; a management request signal is generated according to the state change amount, and the management request signal includes:

[0080] When the parking status change trend shows that the occupancy rate continues to exceed the set threshold, a vehicle dispatch request signal is generated. The signal content includes a dynamic identifier, a parking space occupancy rate change trend, and a vehicle dispatch suggestion.

[0081] When the change trend of the equipment status quantity shows that the frequency of the operating parameters exceeds the preset range, an equipment inspection request signal is generated. The signal content includes the dynamic identifier, equipment operating parameters and fault prediction results;

[0082] When the change trend of the waste state quantity shows that the filling rate growth rate exceeds the set value, a waste cleaning request signal is generated, and the signal content includes a dynamic identifier, the filling growth trend of the waste storage area and a cleaning time suggestion;

[0083] The above management request signal is sent to the management terminal through the remote communication module, triggering the corresponding management measures of the park and synchronously recorded as optimization data records.

[0084] In step 3, the host computer generates corresponding management measures according to the received request signal, and uses the trained management prediction model to output the dynamic identifiers of the target unit area and the N unit areas adjacent to the target unit area after the implementation of the management measures; the park status is optimized by the management measures corresponding to the minimum sum of the weighted sum of the parking status quantity, equipment status quantity and waste status quantity of each unit area.

[0085] Preferably, combining real-time status record data and a time series of unit area dynamic identifiers, a management prediction model is used to analyze the state change trend of the unit area; based on the analysis results, optimization suggestions are generated, including adding new parking spaces, adjusting parking area layouts, optimizing equipment operation configurations, and expanding waste storage capacity; and the optimization suggestions are associated with the unit area dynamic identifiers to generate an optimization data series.

[0086] Preferably, the multi-view images in the park video data are combined, and a three-dimensional reconstruction algorithm is used to perform three-dimensional modeling of the spatial position information, boundary shape and center point coordinates of the unit area based on the dynamic identifier of the unit area to generate a corresponding three-dimensional real scene model; the generated three-dimensional real scene model is subjected to geometric correction and texture mapping;

[0087] Using the dynamic identifiers of unit areas, real-time status data, historical status data, and structural information are dynamically bound to corresponding unit areas in the 3D reality model. Dynamic changes, time series, and spatial distribution of functional data are managed separately through hierarchical organization. Functional data is mapped to corresponding unit areas in the 3D reality model.

[0088] The binding relationship between the generated 3D real scene model and the functional data is stored in a structured format, including the geometric structure of the 3D model, the identifier of the unit area and the recording format of the functional data, providing a standardized data interface.

[0089] Preferably, the method utilizes the dynamic identifier of the unit area to dynamically bind the real-time status data, historical status data and structural information to the corresponding unit area in the three-dimensional real scene model, and manages the dynamic changes, time series and spatial distribution of the functional data in a hierarchical organization manner; mapping the functional data to the corresponding unit area in the three-dimensional real scene model includes:

[0090] Functional data is bound to the 3D real-world model through dynamic rules. For example, the real-time occupancy status of parking spaces is directly bound to the display nodes of the 3D real-world model and updated within seconds. Equipment operating parameters are aggregated by minute and bound to the analysis nodes of the 3D real-world model. Structural information about waste storage areas is bound to the base nodes of the 3D real-world model. All binding logic is adjusted in real time based on data update frequency and functional requirements.

[0091] Functional data is logically separated and managed according to three layers: real-time status, historical data, and structural information. The real-time status, including parking space occupancy status and equipment temperature data, is directly mapped to the dynamic display layer of the 3D real-scene model; historical data, including parking space usage records and equipment operation curves, are bound to the analysis layer of the 3D real-scene model for trend analysis; structural information, including the geometric shape of the unit area and functional classification labels, is bound to the base layer of the 3D real-scene model for maintenance of the 3D real-scene model structure and logical definition of functional areas.

[0092] Preferably, the three-dimensional real scene model of the park is updated based on the optimization data sequence and the real-time status record data, and the dynamic identifier, real-time status and optimization suggestion of the unit area are mapped to the corresponding unit area in the three-dimensional real scene model; the dynamic status, prediction results and optimization suggestion of the unit area are visually presented in the three-dimensional real scene model; the real-time operation status and trend change of each unit area in the park are reflected through the dynamically updated three-dimensional real scene model;

[0093] Furthermore, the generated optimization suggestions and real-time status data are used to construct a three-dimensional real-scene model of the park; through a dynamic mapping method, the dynamic identifier of the unit area is associated with the corresponding unit area in the three-dimensional real-scene model, and the optimization suggestions and real-time status data are embedded in the three-dimensional real-scene model;

[0094] The status display of the three-dimensional real-scene model includes the real-time status of the unit area, optimization suggestions, and visualization of status changes; the current status of the unit area is displayed in real time through dynamic identifiers, including the occupancy status of parking spaces, equipment operating parameters, and the current fill level of the waste storage area; optimization suggestions are displayed in the form of annotations in the corresponding unit area, including parking space layout adjustment plans, equipment operation configuration adjustment plans, and waste storage area expansion plans;

[0095] The 3D real-scene model updates the display content of the unit area in real time based on dynamic mapping rules, including changes in the current status, the implementation progress of optimization suggestions, and the status change trajectory.

[0096] Preferably, the extended visual display of the three-dimensional real scene model includes:

[0097] By mapping dynamic identifiers to unit areas in the 3D reality model, the real-time status of each unit area is combined with future trends. The expanded visualization of parking spaces includes an overlay of a heat map of occupancy during predicted peak hours, combined with color coding to dynamically reflect changes in usage before and after peak hours. The equipment area dynamically overlays vibration frequency and temperature fluctuation trend charts on a timeline, and uses animations to display potential fault risk locations. The scrap storage area dynamically adjusts transparency and overlays a trend curve of future fill levels, displaying the expected full load time node.

[0098] A panoramic view of the 3D real-life model generates a spatial distribution heat map of the predicted status. This heat map combines real-time status, historical data, and future predictions to comprehensively display trends in parking space saturation distribution, equipment abnormality distribution probabilities, and utilization rate changes in waste storage areas. The heat map content is updated in real time using dynamic layer overlays, helping managers understand the park's status and make decisions in real time.

[0099] Preferably, based on the three-dimensional real-scene model and dynamic records, combined with the dynamically updated unit area status data, vehicle scheduling suggestions, equipment maintenance plans and waste disposal plans are sent to the management terminal in real time through the remote communication module; when the status change and prediction results of the unit area are updated, the abnormal area is highlighted in the three-dimensional model in the form of a dynamic mark, and the remote notification content is updated synchronously to trigger the corresponding emergency management mechanism.

[0100] like Figure 2 As shown, embodiment 2 of the present invention provides a three-dimensional smart park status monitoring and optimization system, including: a data acquisition module, a management prediction module, a dynamic mapping model, a status monitoring module, a three-dimensional modeling module, and a communication and management module;

[0101] The data acquisition module is used to acquire video data within the campus coverage area through fixed cameras and UAV dynamic acquisition equipment, and transmit the collected video data to the main control terminal using a wireless communication unit. The main control terminal performs pre-processing operations on the video data, including noise removal, frame rate standardization, and color enhancement.

[0102] Preferably, the data acquisition module includes an image acquisition unit and an image segmentation unit; the image acquisition unit is used to acquire park video data. The image segmentation unit is embedded with an image segmentation model, which is mainly used to extract unit areas from the park video data using the image segmentation model.

[0103] Preferably, the management prediction module is mainly used to determine the type of the unit area, obtain the function of the unit area based on the determined type, and obtain management data of the unit area based on the function.

[0104] The management prediction module is used to identify the type of unit area by analyzing the morphological characteristics of the segmented unit area and combining it with the preset classification standards, and to extract the functional data corresponding to the unit area type from the real-time monitoring data;

[0105] The management prediction module is used to configure a management prediction model based on a BP neural network, analyze pre-processed video data, extract unit areas in the park video, and generate dynamic identifiers. The dynamic identifiers include the center point coordinates, boundary coordinates, and state changes of the unit area. The state changes include the occupancy status of parking spaces, the operating status of equipment, and the filling status of waste storage areas.

[0106] The dynamic mapping module is used to bind the dynamic identifier of the unit area with the video frame sequence based on the dynamic mapping method, and to construct a time series of the dynamic identifier of the unit area, and to record the position change, boundary adjustment and dynamic state change of each unit area in the video frame sequence;

[0107] A status monitoring module is used to determine the status of a unit area in real time by combining the time series of the unit area dynamic identifier and the amount of state change, and to generate a management plan based on the amount of state change. The management plan includes vehicle scheduling suggestions, equipment maintenance plans, and waste disposal plans, and is associated with the time series to generate real-time status record data;

[0108] The 3D modeling module is used to update the park's 3D real-world model based on real-time status record data and optimization suggestions. It also maps the dynamic identifiers, real-time status, and optimization suggestions of unit areas into the 3D real-world model. Dynamic identification highlights abnormal areas and supports the visualization of management plans.

[0109] The communication and management module is used to send vehicle scheduling suggestions, equipment maintenance plans and waste disposal plans to the management terminal in real time through the remote communication module, and synchronously update the dynamic record data when the status changes and prediction results of the unit area are updated, triggering the corresponding emergency management mechanism.

[0110] For example, if the unit area is a parking unit, smart sensors installed on the parking spaces acquire the status and location information (i.e., management data) of each parking space. These smart sensors can be infrared sensors, which detect whether a vehicle is parked in the space. This parking space status and location information is then sent to the unit area classification and functional data extraction module. When a vehicle parked in a parking space has exceeded its time limit or a parking space has other faults, the management prediction and anomaly analysis module uses the acquired fault information to render and visualize the corresponding unit area in the 3D real-world model, using flashing, different colors, and other methods, so that managers can obtain the fault location and information in real time. At this point, the remote notification and emergency management module generates a remote management notification message, which is received by the corresponding manager's terminal (e.g., mobile phone, tablet, desktop computer, etc.). The manager is then required to manage the corresponding park project, such as troubleshooting, within a set timeframe after receiving the remote management notification message.

[0111] For another example, if the unit area is a waste storage unit in a chemical park, the waste capacity of each waste hopper in the waste unit can be obtained through sensors installed on the waste hoppers, and this data is sent to the data acquisition module. The unit area classification and functional data extraction module then uses the management prediction model to predict the time when the waste hopper will be filled based on the current waste capacity, and the predicted time is rendered and visualized digitally, in different colors, etc. on the corresponding unit area in the three-dimensional real-life model. At this time, the remote notification and emergency management module generates remote management notification information, which is received by the corresponding manager's terminal (such as a mobile phone, tablet computer, desktop computer, etc.). After receiving the remote management notification information, the manager needs to implement the management of the corresponding project in the park according to the predicted time, such as clearing the chemical waste in the waste hopper.

[0112] Furthermore, in order to prevent managers from missing this information within the predicted time, when the predicted time is reached, this information is rendered again in the three-dimensional real-scene model and remotely notified.

[0113] In another exemplary embodiment of the present application, a first wireless communication unit may be provided in the data acquisition and preprocessing module to communicate with monitoring equipment or drones in the park and obtain park video data.

[0114] Among them, when the data acquisition module obtains video data through the monitoring equipment in the park, this application also needs to set up an image data fusion module to fuse the video data obtained by all monitoring devices based on the original image blueprint to obtain an overall image of the park.

[0115] In another exemplary embodiment of the present application, the data acquisition module, the three-dimensional module, the management prediction module and the remote notification management module can be configured as an integrated circuit. In this case, the entire system provided by the present application is equivalent to a processor or a host computer.

[0116] In another exemplary embodiment of the present application, a second wireless communication unit may be provided in the data collection module, and the second wireless communication unit is used to communicate with a data collection device in the unit area to obtain management data of the unit area.

[0117] In another exemplary embodiment of the present application, in order to monitor in real time whether the managers of the corresponding unit area are able to operate on the projects to be managed, or to detect whether the remote communication link of the remote notification and emergency management module fails, so that the managers can perform corresponding operations in a timely manner, a display monitoring unit and an alarm unit can be set in the management prediction and abnormality analysis module.

[0118] When the management prediction result does not meet the set conditions, the management prediction result will be rendered and visualized on the three-dimensional real-scene model. The display monitoring unit will obtain the display time of the management prediction result in real time. When the display time exceeds the preset time range, an alarm instruction will be generated.

[0119] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0120] Those skilled in the art will appreciate that all or part of the processes in the above embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0121] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0122] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for monitoring and optimizing the status of a three-dimensional smart park, wherein the park's status monitoring objects include parking spaces, equipment, and waste storage areas, wherein: The equipment includes new energy equipment and traditional energy equipment; and is characterized by including: Step 1, obtain the park image video; establish a management prediction model based on the BP neural network; use the trained management prediction model to extract the unit area from the park image video and assign a dynamic identifier to each unit area, the dynamic identifier includes: the coordinates of the center point of the unit area, the coordinates of the boundary of the unit area and the state change of the unit area, wherein the state change of the unit area includes the parking state quantity, the equipment state quantity, and the waste state quantity; when the equipment in the unit area is operating normally, the equipment state quantity is set to 1, otherwise the equipment state quantity is set to 0; when the equipment state quantity is 1 and the parking spaces in the unit area are occupied by 70% or more, or the equipment state quantity is 0 and the parking spaces in the unit area are occupied by 50% or more, the parking state quantity is set to 1, otherwise the parking state quantity is set to 0; when the equipment state quantity is 1 and the filling quantity of the waste storage area in the unit area is 70% or more, or the equipment state quantity is 0 and the filling quantity of the waste storage area in the unit area is 50% or more, the waste state quantity is set to 1, otherwise the waste state quantity is set to 0; Based on the dynamic mapping method, the unit area and the dynamic identifier of each unit area corresponding to each frame in the frame sequence of the park image video are determined, and the time sequence of the dynamic identifier of each unit area is constructed according to the time sequence of each frame; The dynamic identifier includes geometric information and dynamic status information of the unit area, wherein the geometric information includes the center point coordinates and boundary coordinates of the unit area, and the dynamic status information includes the real-time occupancy status of the parking space, the operating status of the equipment, and the filling status of the waste storage area; Dynamic identifiers are generated by analyzing the morphological features and spatial distribution of unit areas in video frames. Parking spaces are identified by regularly arranged rectangular areas, equipment areas by morphological features with regular boundaries, and waste storage areas by dynamically changing boundary contours and fill ratios. Dynamically map the dynamic identifier of each unit area with the video frame sequence, and record the position change, boundary adjustment and state change information of each unit area in the video; The dynamic mapping of the dynamic identifier of each unit area with the video frame sequence, and recording of the position change, boundary adjustment and state change information of each unit area in the video, includes: In continuous video frames, the unit area is tracked across frames according to dynamic mapping rules. Position continuity rules are set to limit the offset of the center point coordinates to no more than five pixels. Boundary morphological similarity rules are set to require the boundary overlap rate to reach more than 70 percent. The dynamic identifiers of the unit area are combined to match the status data of parking spaces, equipment areas, and waste storage areas. The geometric change information of the unit area is recorded through cross-frame tracking, including the dynamic adjustment of the center point coordinates and the shape change of the boundary contour; the state change amount of the unit area is recorded. When the equipment in the unit area is operating normally, the equipment state amount is set to 1; when the equipment stops abnormally, the equipment state amount is set to 0; when the equipment state amount is 1 and the occupancy ratio of parking spaces in the unit area reaches 70% or above, or the equipment state amount is 0 and the occupancy ratio of parking spaces reaches 50% or above, the parking state amount is set to 1, otherwise it is set to 0; when the equipment state amount is 1 and the filling amount of the waste storage area in the unit area reaches 70% or above, or the equipment state amount is 0 and the filling amount of the waste storage area reaches 50% or above, the waste state amount is set to 1, otherwise it is set to 0; the state change amount of the unit area is used to dynamically reflect the changes in the parking, equipment and waste storage states of the unit area; Integrate position changes, boundary outline updates, and dynamic state transitions into time series along the time axis; Step 2: Extract the dynamic identifier of the target unit area during the monitoring period from the time series of the dynamic identifiers of each unit area; if the parking state variable of the extracted dynamic identifier is 1, generate a vehicle dispatch request signal for the target unit area; if the equipment state variable is 1, generate an equipment inspection request signal for the target unit area; if the waste state variable is 1, generate a waste removal request signal for the target unit area; if the parking state variable, equipment state variable, and waste state variable are all 1, generate an alarm signal for the target unit area; In step 3, the host computer generates corresponding management measures according to the received request signal, and uses the trained management prediction model to output the dynamic identifiers of the target unit area and the N unit areas adjacent to the target unit area after the implementation of the management measures; the park status is optimized by the management measures corresponding to the minimum sum of the weighted sum of the parking status quantity, equipment status quantity and waste status quantity of each unit area.

2. The method for monitoring and optimizing the status of a three-dimensional smart park according to claim 1, characterized in that: Step 1 includes: Fixed cameras and drone-based dynamic data acquisition equipment deployed throughout the park capture real-time video data covering the entire park. Wireless communication units are used to transmit the video data to the main control terminal, which performs pre-processing operations on the video data, including noise removal, frame rate standardization, and color enhancement. Based on parking space occupancy records, equipment operation logs, and historical management data from waste storage areas, core features were extracted and labeled to construct a data set for training a management prediction model using a BP neural network. The labeled data set was trained using the BP neural network to generate a management prediction model. The classification capability of the management prediction model was optimized using a random forest classification model, and the weight parameters of the management prediction model were adjusted through cross-validation. The image and video data are analyzed frame by frame, and the trained management prediction model is applied to identify the unit areas in the park. The unit areas include parking spaces, equipment areas, and waste storage areas. For each identified unit area, a dynamic identifier corresponding to the dynamic state is generated. The dynamic identifier includes the coordinates of the center point of the unit area, the coordinates of the boundary of the unit area, and the state change of the unit area.

3. The method for monitoring and optimizing the status of a three-dimensional smart park according to claim 1, characterized in that: Step 2 includes: Analyze the state change trend of the target unit area based on the unit area dynamic identifier time series; Generate a real-time status record based on the dynamic changes of the parking state quantity, the equipment state quantity and the waste state quantity, wherein the real-time status record includes a state change curve and anomaly detection results of the target unit area during the monitoring period; Anomaly detection results are generated by analyzing the behavior of exceeding the limit of parking space occupancy time, the deviation of equipment operating parameters from the normal range, and the continuous growth trend of the filling rate of the waste storage area; The real-time status records are bound to the time series of dynamic identifiers through dynamic mapping method.

4. The method for monitoring and optimizing the status of a three-dimensional smart park according to claim 3, characterized in that: The real-time status record includes a dynamic identifier of the target unit area and a state change amount; Generate a management request signal based on the state change. The management request signal includes: When the parking status change trend shows that the occupancy rate continues to exceed the set threshold, a vehicle dispatch request signal is generated. The signal content includes a dynamic identifier, a parking space occupancy rate change trend, and a vehicle dispatch suggestion. When the change trend of the equipment status quantity shows that the frequency of the operating parameters exceeds the preset range, an equipment inspection request signal is generated. The signal content includes the dynamic identifier, equipment operating parameters and fault prediction results; When the change trend of the waste state quantity shows that the filling rate growth rate exceeds the set value, a waste cleaning request signal is generated, and the signal content includes a dynamic identifier, the filling growth trend of the waste storage area and a cleaning time suggestion; The above management request signal is sent to the management terminal through the remote communication module, triggering the corresponding management measures of the park and synchronously recorded as optimization data records.

5. The method for monitoring and optimizing the status of a three-dimensional smart park according to claim 1, characterized in that: Step 3 includes: After receiving the request signal from the target unit area, the host computer extracts the real-time status data and historical time series data of the target unit area and its N adjacent unit areas through the management prediction model, and uses multi-objective optimization to calculate the change trends of the parking state quantity, equipment state quantity, and waste state quantity respectively; Based on the trend calculation results, a dynamic weighting strategy is used to adjust the state quantity weight. The parking state quantity weight is adjusted according to the priority of vehicle scheduling needs. The equipment state quantity weight changes dynamically according to the equipment operation abnormality rate. The waste state quantity weight is calculated based on the filling growth rate. During the optimization calculation process, the state quantities of adjacent unit areas and the target unit area are analyzed collaboratively, giving priority to the feasibility of vehicle scheduling due to parking space saturation, the coverage of equipment inspections, and the time and resource constraints of waste removal paths. Generate a resource allocation plan for the target unit area and its adjacent areas through a state quantity weighted sum minimization optimization model. The resource allocation plan includes vehicle scheduling target area allocation, equipment inspection route optimization, and waste cleanup time scheduling. The resource allocation plan is bound to the dynamic identifier of the unit area, the state change in the dynamic identifier is updated, and an optimized dynamic identifier sequence is generated to optimize the park state.

6. The method for monitoring and optimizing the status of a three-dimensional smart park according to claim 5, characterized in that: The generation of the resource allocation plan includes: For parking resource allocation, the shortest path method is used to plan vehicle dispatch routes based on the dynamic identifiers of the target unit area and its adjacent areas, combined with vehicle flow trends and real-time parking space occupancy rates calculated from historical monitoring data. Path optimization takes into account the distance from the vehicle's current location to the target parking area, the parking space availability rate, and the priority of parking needs. For resource allocation of equipment inspections, we utilize time series data of equipment operating parameters, combined with equipment operating anomaly rates, to generate an inspection plan through an inspection path optimization algorithm. This inspection plan prioritizes the degree of equipment anomalies and the path with the shortest inspection time. For waste disposal resource allocation, the priority and arrival order of disposal vehicles are assigned based on the waste storage area's fill rate growth trend and the disposal time window. The disposal route optimization is based on the waste storage area's full load time prediction, the disposal vehicle's operating capacity, and the accessibility of the logistics channel. The resource allocation scheme updates the state change of the target unit area and its adjacent areas through dynamic identifiers, generates an optimized dynamic identifier sequence and applies it.

7. The method for monitoring and optimizing the status of a three-dimensional smart park according to claim 6, characterized in that: The optimized dynamic identifier sequence and application includes: Using an optimized dynamic identifier sequence, the status display of unit areas in the park's 3D model is updated in real time. Parking status is dynamically displayed in color-coded form, equipment status is displayed with boundary highlights and operating parameter curves superimposed, and waste status is dynamically displayed with a fill transparency gradient. By linking resource allocation plans with dynamic identifiers of corresponding unit areas in the 3D model through multi-dimensional structured data, the execution plan for vehicle dispatch, equipment inspection, and waste disposal is visualized, allowing managers to intuitively view status changes and optimization measures for each unit area. When the state change amount in the dynamic identifier sequence triggers the preset threshold, a new management request signal is generated in combination with the real-time state data to start the next round of resource optimization and state adjustment.

Images