Smart scenic area zoning and shunting method based on tourist flow prediction and dynamic bearing control

Abstract

The invention discloses a smart scenic area zoning and shunting method based on tourist flow prediction and dynamic bearing control, and relates to the technical field of smart scenic area passenger flow prediction, and the method comprises the steps: collecting historical tourist flow data and real-time tourist flow data, constructing a tourist flow prediction model, and generating a tourist flow prediction value; calculating a reference bearing threshold value, dynamically adjusting a current bearing threshold value in combination with the tourist flow predicted value, and when the tourist flow predicted value is greater than a preset proportion of the adjusted threshold value, marking the tourist flow predicted value as an early warning subarea; constructing a tourist flow transfer matrix between the early warning subarea and the adjacent subarea, calculating a tourist flow transfer probability value, and obtaining a shunting target subarea; and pushing the guide information of the shunting target zone to the early warning zone tourist terminal, synchronously adjusting the entrance gate of the early warning zone to reduce the passing rate, and improving the passing rate of the entrance gate of the shunting target zone. According to the invention, the execution success rate and tourist acceptability of the shunting scheme are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent scenic spot passenger flow prediction, and particularly to an intelligent scenic spot partition flow distribution method based on tourist flow prediction and dynamic bearing control. BACKGROUND

[0002] With the rapid development of intelligent tourism and digital scenic spot construction, the management mode of scenic spots is gradually changing from traditional manual monitoring and static flow control to a dynamic management mode based on data driving. At present, most scenic spots have introduced video monitoring, mobile signaling, ticket reservation systems and other data sources to realize tourist number monitoring and trend prediction through a passenger flow analysis platform. However, in the practice of scenic spot tourist flow control, the following problems still exist: first, most existing prediction models are based on a single historical flow sequence or static period law, and fail to fully introduce multi-factor comprehensive correction such as holidays and weather, so the prediction accuracy is limited; second, most flow control strategies use fixed thresholds, and it is difficult to dynamically adjust the bearing capacity in combination with real-time tourist distribution and safety evacuation conditions; third, after the tourist density exceeds the standard, it usually relies on broadcast prompts or manual intervention for diversion, and lacks an automatic dynamic flow distribution control mechanism at the partition level. These deficiencies lead to problems such as local area congestion, declining tourist experience, and rising safety risks in scenic spots during holiday peak periods. Therefore, there is an urgent need for a comprehensive flow distribution method that combines tourist flow prediction, dynamic bearing control and partition guidance strategy to realize real-time regulation and distribution optimization of tourists in intelligent scenic spots.

[0003] CN118134040A discloses a scenic spot tourist number control method, device, equipment and medium. The method obtains city data information and user signaling data of a target city, constructs a key grid map based on the city data information, and uses a time series model to predict and process the tourist number of the key grid area, so as to determine a tourist number control strategy according to the prediction result and target tourist bearing capacity parameters of each tourist attraction. This scheme realizes prediction and control of tourist number at the city level, can alleviate the problem of tourist concentration in popular areas during the tourist peak period, and improves the tourist experience. However, the prediction granularity of this method still stays at the city or attraction level, and does not establish a partition-level prediction mechanism for each functional partition in the scenic spot. The bearing control parameters are statically set, and lack the ability to dynamically adjust the threshold in combination with the safety evacuation conditions. Moreover, it does not construct a tourist flow probability model between different partitions, and it is difficult to realize active flow distribution and guidance of overloaded partitions.

[0004] CN109413129A discloses an intelligent tourist diversion system for urban tourism, comprising a data platform, scenic area servers, municipal transportation departments, and user terminals. When tourists request ticket purchases, the system uses multi-step analysis through the data center to dynamically monitor tourist density across scenic areas within the city. When scenic area visitor flow is high, it sends a relief request to the municipal transportation department to achieve tourist diversion and traffic management between scenic areas. This invention achieves balanced tourist distribution and information linkage at the cross-scenic area level, exhibiting a certain degree of real-time performance and coordination. However, this solution primarily focuses on macro-level control between scenic areas or at the city level, failing to delve into the subdivided areas within a single scenic area for prediction and diversion. Furthermore, its diversion strategy relies on external traffic coordination, lacking an adaptive diversion mechanism based on changes in the scenic area's carrying capacity, and is insufficient in addressing congestion or potential safety risks within specific scenic areas.

[0005] In summary, existing tourist flow prediction and diversion control technologies for scenic areas suffer from limited prediction accuracy, static threshold adjustments, and a single diversion mechanism, making it difficult to meet the needs of smart scenic areas for "zoning-level, dynamic, and closed-loop" tourist management. This invention proposes a zoned diversion method for smart scenic areas based on tourist flow prediction and dynamic capacity control. This scheme achieves predictive control and adaptive diversion adjustment of tourist flow in scenic areas, effectively alleviating local congestion and improving tourist safety and experience. Summary of the Invention

[0006] The purpose of this section is to outline some aspects of the embodiments of the present invention and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this section, the abstract and title of the invention. Such simplifications or omissions shall not be used to limit the scope of the present invention.

[0007] In view of the problems of insufficient prediction accuracy, static carrying capacity threshold, and lack of real-time adaptability of the diversion mechanism in existing tourist flow prediction and diversion control technologies, this invention is proposed.

[0008] Therefore, the problem to be solved by this invention is how to achieve coordinated control of multi-factor tourist flow prediction, dynamic carrying capacity threshold adjustment and inter-zone flow diversion at the scenic area zoning level, so as to achieve balanced distribution and guidance optimization of tourists in different zones while ensuring safe evacuation capacity.

[0009] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide a smart scenic area zoned flow control method based on tourist flow prediction and dynamic carrying capacity control, comprising, Collect historical and real-time visitor flow data for each zone of the scenic area, construct a visitor flow prediction model, and generate predicted visitor flow values ​​for each zone for future periods. The baseline carrying capacity threshold is calculated based on the spatial capacity parameters and safe evacuation parameters of each zone. The current carrying capacity threshold is then dynamically adjusted in conjunction with the predicted tourist flow to obtain the warning zones. Construct a tourist flow transfer matrix between the warning zone and adjacent zones, calculate the tourist flow transfer probability value, and obtain the diversion target zone; The system pushes guidance information for the diversion target zone to the visitor terminal in the warning zone, and simultaneously adjusts the entrance gate of the warning zone to reduce the passage speed and increases the passage speed of the entrance gate of the diversion target zone.

[0010] Secondly, embodiments of the present invention provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any step of the above-mentioned smart scenic area zoning and flow control method based on tourist flow prediction and dynamic carrying capacity control.

[0011] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the above-described smart scenic area zoning and flow control method based on tourist flow prediction and dynamic carrying capacity control.

[0012] Compared with existing technologies, the advantages of this invention are as follows: A tourist flow prediction model is constructed based on LSTM neural networks, fusing historical data, holiday factors, and meteorological parameters, providing a 30-minute to 2-hour advance prediction window for management decisions, thus establishing a time advantage from the outset; By dynamically adjusting the carrying capacity threshold and introducing a carrying capacity pressure index, the limitations of traditional fixed thresholds are overcome, achieving a more precise and adaptive early warning mechanism, effectively avoiding false alarms and missed alarms; By utilizing a tourist flow transfer matrix to comprehensively quantify distance weights and landscape attractiveness weights, the selection of diversion targets is transformed from experience-based judgment to data-driven scientific decision-making, significantly improving the success rate of diversion schemes and tourist acceptance; Through a combination of information push and gate speed adjustment, a dual synergy of flexible guidance and rigid control is achieved, ensuring rapid response and precise execution of diversion measures, maximizing the scenic area's carrying capacity and service quality while ensuring tourist safety. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart of a smart scenic area zone-based flow control method based on tourist flow prediction and dynamic carrying capacity control. Detailed Implementation

[0014] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0015] Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort should fall within the scope of protection of this invention.

[0016] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0017] As mentioned in the background section, existing tourist flow prediction and diversion control technologies for scenic areas suffer from limited prediction accuracy, static threshold adjustments, and a single diversion mechanism, making it difficult to meet the needs of smart scenic areas for "zoning-level, dynamic, and closed-loop" tourist management. To address these issues, this invention provides a zoned diversion method for smart scenic areas based on tourist flow prediction and dynamic capacity control.

[0018] Reference Figure 1 , Figure 1 This is a flowchart illustrating a smart scenic area zoning and flow control method based on tourist flow prediction and dynamic carrying capacity control, according to an embodiment of the present invention. Figure 1 As shown, a smart scenic area zoning and flow control method based on tourist flow prediction and dynamic carrying capacity control includes: S1: Collect historical and real-time visitor flow data for each zone of the scenic area, construct a visitor flow prediction model, and generate predicted visitor flow values ​​for each zone in future time periods; Specifically, visitor flow statistics equipment and meteorological monitoring equipment are deployed at the entrances of each zone and key passage nodes of the scenic area; the visitor flow statistics equipment is equipped with a binocular visitor flow counter and a thermal imaging sensor; the meteorological monitoring equipment is equipped with a temperature sensor, a humidity sensor and a light intensity sensor, and multiple devices use a unified clock synchronization mechanism for data collection; the visitor flow statistics equipment and the meteorological monitoring equipment are connected to the scenic area's data center server through a wireless network transmission module.

[0019] S1.1: Collect the number of visitors entering and exiting each zone of the scenic area over a continuous period of time through visitor flow statistics equipment, classify and store the data according to zone identifier and timestamp, and obtain historical visitor flow data; It should be noted that the historical visitor flow data includes hourly visitor numbers, daily cumulative visitor numbers, weekly average visitor numbers, and monthly visitor distribution information for each zone; the timestamp records the year, month, day, hour, minute, and second information of the time of collection; and the zone identifier is used to distinguish different functional areas within the scenic area.

[0020] For example, when monitoring visitor flow at a 5A-level scenic area, staff installed binocular visitor counters at the main entrance, core scenic area A, ecological trail area B, and leisure plaza area C. The equipment counted the number of people passing through every minute. In the past year of monitoring, historical visitor flow data for core scenic area A showed that the average daily visitor volume on weekdays was approximately one value, the average daily visitor volume on weekends was approximately another value, and the highest single-day visitor volume reached the peak value during the Spring Festival Golden Week. Due to its relatively remote terrain, historical data for ecological trail area B showed that the average daily visitor volume was relatively low. The specific timestamp (e.g., year, month, day, hour, minute, second) and corresponding zone identifier (e.g., "core scenic area A") of each data point were recorded.

[0021] S1.2: Collect real-time data on the number of visitors entering and exiting each zone of the scenic area at the current moment, and match and associate it with historical visitor flow data according to the same zone identifier and time format to obtain real-time visitor flow data; It should be noted that real-time visitor flow data includes the instantaneous number of visitors in each zone at the current moment, the visitor growth rate in the past hour, and the cumulative number of visitors for the day; the matching association is to establish a time-series correspondence between historical trends and the current state.

[0022] S1.3: Obtain holiday date information for the area where the scenic spot is located, calculate the holiday impact weight coefficient based on the holiday type and duration, and construct the holiday factor vector; Specifically, a list of holiday dates is extracted from the national statutory holiday schedule notice, and the holiday types are divided into national statutory holiday categories, local characteristic holiday categories, and weekend categories. Based on the duration of the holiday and the time interval from the current date, an exponential decay function is used to calculate the holiday impact weight coefficient. The holiday impact weight coefficient increases with the duration of the holiday and decreases with the time interval from the current date.

[0023] Preferably, the holiday factor vector is indexed by date and includes a holiday type identifier, a holiday impact weight coefficient, and identifiers of the preceding and following workdays; the holiday type identifier is represented by one-hot encoding, and the holiday impact weight coefficient ranges from zero to a certain positive number.

[0024] S1.4: Collect real-time meteorological data by using meteorological monitoring equipment to collect current temperature, humidity, rainfall, and visibility data of the scenic area; It should be noted that real-time meteorological data is recorded in the same timestamp format as real-time tourist flow data to establish a time-series synchronization relationship between meteorological conditions and tourist flow.

[0025] S1.5: Perform statistical regression analysis on the meteorological data and tourist flow values ​​for the corresponding dates in the historical tourist flow data, calculate the impact coefficient of different meteorological conditions on tourist flow, and obtain weather correction parameters; Specifically, the average daily temperature, average humidity, cumulative rainfall, and average visibility data were extracted from historical tourist flow data. These average values ​​were used as independent variables, and the tourist flow data for the corresponding date was used as the dependent variable. A multiple linear regression method was used to fit the quantitative relationship between meteorological elements and tourist flow, and the regression coefficients of each meteorological element were obtained as weather correction parameters.

[0026] Preferably, the weather correction parameters include temperature influence coefficient, humidity influence coefficient, rainfall influence coefficient and visibility influence coefficient; when the temperature data is within the temperature range suitable for travel, the temperature influence coefficient is positive; when the rainfall data exceeds the rainfall warning threshold, the rainfall influence coefficient is negative and its absolute value increases with the increase of rainfall intensity.

[0027] It should be noted that the weather correction parameters are obtained through repeated iterative fitting of historical data, which can quantify the impact of different weather conditions on tourists' willingness to travel and the carrying capacity of scenic spots.

[0028] S1.6: Divide the historical tourist flow data into training dataset and validation dataset according to time order, and extract the time feature variables and spatial feature variables from the training dataset; Specifically, hourly, weekday, and monthly serial numbers are extracted from historical tourist flow data as time feature variables; geographical coordinates, area parameters, and attraction type identifiers of each zone are extracted as spatial feature variables; the holiday type identifiers and holiday impact weight coefficients of the corresponding dates in the holiday factor vector are added to the time feature variables; and temperature, humidity, rainfall, and visibility data corresponding to the weather correction parameters are added to the time feature variables to construct a training sample matrix containing multidimensional features.

[0029] It should be noted that the training sample matrix uses timestamps as row indices and time feature variables, spatial feature variables, holiday factor vectors, and weather correction parameters as column indices. The number of columns in the training sample matrix is ​​the sum of all feature dimensions.

[0030] S1.7: Construct a tourist flow prediction model based on a long short-term memory neural network, input the training sample matrix into the tourist flow prediction model for training, and iteratively optimize the model parameters through a validation dataset; Specifically, the tourist flow prediction model includes an input layer, a long short-term memory hidden layer, and an output layer. The input layer receives multi-dimensional feature data from the training sample matrix. The long short-term memory hidden layer contains forget gate units, input gate units, and output gate units, which capture the temporal dependencies in historical tourist flow data through a gating mechanism. The output layer generates predicted tourist flow values ​​for future periods.

[0031] Preferably, a holiday attention mechanism submodule and a weather attention mechanism submodule are set in the long short-term memory hidden layer; the holiday attention mechanism submodule dynamically adjusts the feature weights according to the holiday influence weight coefficient in the holiday factor vector; the weather attention mechanism submodule dynamically adjusts the contribution of meteorological features according to the temperature influence coefficient, humidity influence coefficient, rainfall influence coefficient and visibility influence coefficient in the weather correction parameters.

[0032] It should be noted that the forget gate unit determines the degree of retention of historical information, the input gate unit controls the amount of new information written, and the output gate unit determines the output value at the current moment; through the gating mechanism, the tourist flow prediction model can adaptively learn long-term trends and short-term fluctuations.

[0033] S1.8: Input the real-time tourist flow data, the holiday factor vector corresponding to the current date, and the real-time weather data into the trained tourist flow prediction model to generate the tourist flow prediction value; Specifically, the formula for predicting tourist flow is as follows: ; in, Here is the predicted tourist flow for partition p in time period t. It is the Sigmoid activation function. Historical traffic weighting coefficient This represents the LSTM hidden state for the first i time periods. This represents the impact coefficient of holidays. Single-hot encoding for holiday types, This is a meteorological correction factor. For meteorological parameter vectors, The current temperature. For optimal temperature during visit, For rainfall, As a temperature regulation factor, is the rainfall adjustment factor, n is the number of historical periods, q is the meteorological parameter dimension, and m is the holiday characteristic dimension.

[0034] Preferably, the tourist flow prediction model calculates the current tourist flow trend based on real-time tourist flow data, judges the impact intensity of holidays in future periods by combining holiday factor vectors, and corrects the prediction results based on real-time meteorological data and weather correction parameters; the tourist flow prediction value is output in units of zones according to future continuous periods, and each period includes the average number of predicted tourists and the prediction confidence interval.

[0035] It should be noted that the tourist flow forecast covers the tourist inflow, outflow, and number of tourists staying in each zone within the next few hours; the forecast confidence interval represents the range of uncertainty of the forecast result; the forecast confidence interval expands as the forecast period lengthens.

[0036] In an optional embodiment, the predicted tourist flow value is compared and analyzed with the actual tourist flow data to calculate the prediction error index. When the prediction error index exceeds the preset error threshold, the model parameter adaptive update mechanism is activated, and the tourist flow prediction model is incrementally trained using the latest collected real-time tourist flow data. The holiday attention mechanism submodule and the weather attention mechanism submodule adjust the attention weight allocation according to the error feedback to improve the tourist flow prediction model's responsiveness to sudden changes in tourist flow.

[0037] S2: Calculate the baseline carrying capacity threshold based on the spatial capacity parameters and safe evacuation parameters of each zone, and dynamically adjust the current carrying capacity threshold in combination with the predicted tourist flow to obtain the early warning zone; Specifically, a zone carrying capacity assessment module and a dynamic early warning judgment module are established in the scenic area's data center server. The zone carrying capacity assessment module is configured with a geographic information database and a safety specification parameter library. The dynamic early warning judgment module is configured with a threshold calculation engine and an early warning trigger. The two modules interact in real time through a data bus. The zone carrying capacity assessment module transmits the calculation results to the dynamic early warning judgment module for early warning status determination.

[0038] S2.1: Obtain topographic mapping data for each zone of the scenic area, extract the activity area, main passage width, and number of entrances and exits for each zone, and construct a set of spatial capacity parameters. Specifically, the boundary coordinate information of each zone is extracted through the scenic area planning map and on-site survey report. The usable area within each zone is calculated after deducting the building footprint, water area, and restricted areas. The road width of the main tourist routes within each zone is measured and recorded. The number of all access routes connecting each zone to the external area is counted to obtain the number of entrances and exits.

[0039] It should be noted that the set of space capacity parameters is indexed by the zone identifier and includes values ​​for the usable area, the width of the main passageway, and the number of entrances and exits.

[0040] For example, when calculating the spatial capacity of a mountainous scenic area, the total area of ​​the core viewing platform A area, as shown by topographic survey data, is a certain number of square meters. After deducting the space occupied by facilities such as viewing platform railings, rest seats, and signs, the usable area is approximately another number of square meters. The main hiking trail in this area is a certain number of meters wide and has a certain number of entrances and exits connecting the uphill trail and the downhill cable car station. Due to the large topographic relief, the usable area of ​​the ecological forest area B area is relatively small. The main passage is a wooden boardwalk, with a main passage width of approximately a certain number of meters, and only a certain number of entrances and exits.

[0041] S2.2: Set the standard value of per capita area based on the terrain features and tour methods of each zone, and calculate the spatial carrying capacity value of each zone by dividing the value of the usable area by the standard value of per capita area. Specifically, according to the National Tourist Attraction Safety Management Standards, the zones are divided into open flat land, mountain trails, indoor venues, and water sightseeing areas. For open flat land zones, the standard value for per capita area is set as the first standard value. For mountain trail zones, due to terrain limitations and safety distance requirements, the standard value for per capita area is set as the second standard value, which is greater than the first standard value. For indoor venue zones and water sightseeing zones, corresponding standard values ​​for per capita area are set according to fire safety regulations and water safety regulations, respectively.

[0042] It should be noted that the space carrying capacity value represents the maximum number of tourists that can be accommodated in a zone at the same time, under the premise of ensuring the comfort and safety of tourists; the standard value of per capita area is dynamically adjusted according to the zone function and visitor density requirements.

[0043] S2.3: Obtain the total length data of evacuation routes and emergency evacuation time limits for each zone. Calculate the safe evacuation capacity of each zone based on the number of entrances and exits, the width of main passages, and the total length data of evacuation routes. Specifically, the emergency evacuation time limits set for each zone are extracted from the scenic area's emergency plan document. These limits represent the maximum allowable time to evacuate all tourists within a zone to a safe area in the event of an emergency. The lengths of all evacuation routes within each zone are measured and summed to obtain the total length of the evacuation routes. Based on the crowd evacuation dynamics model, combined with the width of the main routes, the number of entrances and exits, and the emergency evacuation time limits, the maximum flow of people passing through the exits per unit time is calculated to obtain the safe evacuation capacity.

[0044] The preferred formula for calculating the safe evacuation capacity is as follows: ; in, For safe evacuation capacity values, The number of evacuation routes, The width of channel i To achieve the maximum evacuation speed, For emergency evacuation time limits, Let i be the length of channel i. At normal walking speed, For the area of ​​the safe zone, The area occupied per person, This refers to the export volume.

[0045] It should be noted that the safe evacuation capacity value reflects the maximum safe carrying capacity of a zone under emergency conditions; the evacuation rate per unit width parameter is adjusted according to the channel type and slope conditions, and the evacuation rate of flat and wide channels is higher than that of narrow and steep channels.

[0046] S2.4: Compare the spatial carrying capacity value and the safe evacuation capacity value, and select the smaller one as the benchmark carrying capacity threshold for each zone; Specifically, the system iterates through all zones of the scenic area, extracts the spatial carrying capacity value and the safe evacuation capacity value corresponding to each zone, and performs a value comparison operation. When the spatial carrying capacity value is less than the safe evacuation capacity value, the spatial carrying capacity value is assigned as the baseline carrying capacity threshold. When the safe evacuation capacity value is less than or equal to the spatial carrying capacity value, the safe evacuation capacity value is assigned as the baseline carrying capacity threshold.

[0047] It should be noted that the baseline carrying capacity threshold represents the theoretical maximum carrying capacity of each zone under normal operating conditions, while simultaneously meeting the requirements for spatial comfort and safe evacuation. The minimum value principle is adopted to ensure that the carrying capacity of a zone is limited by the shortcomings in both spatial and safety dimensions.

[0048] Preferably, a baseline carrying capacity threshold data table is established, with the partition identifier as the primary key, to store the baseline carrying capacity threshold, spatial carrying capacity value, safe evacuation capacity value and its calculation timestamp for each partition, forming a basic database of carrying capacity.

[0049] S2.5: Extract the predicted tourist flow value and obtain the predicted tourist number sequence for each partition in future consecutive time periods; Specifically, the predicted tourist flow values ​​for each zone in the future are read from the output of the tourist flow prediction model. The predicted tourist flow values ​​include the prediction period identifier, the zone identifier, and the average number of predicted tourists. The predicted data for the same zone are sorted according to the time order of the prediction period identifier to construct a sequence of predicted tourist numbers.

[0050] It should be noted that the predicted tourist number sequence is a time-ordered numerical array, with each array element representing the number of tourists expected to be received by a zone within the corresponding time period; the time period identifier is consistent with the time period division of the tourist flow prediction value in step S1.

[0051] S2.6: Calculate the dynamic adjustment coefficient based on the numerical trend of the predicted tourist number sequence and real-time meteorological data; Specifically, the baseline carrying capacity threshold is multiplied by a dynamic adjustment coefficient to obtain the current carrying capacity threshold for each zone in the corresponding time period; the numerical growth rate of adjacent time periods in the predicted tourist number sequence is analyzed. When the growth rate continues to rise, it indicates that the tourist flow is in a rapid growth phase. At this time, the dynamic adjustment coefficient is reduced to tighten the carrying capacity limit in advance; when the growth rate tends to stabilize or decline, the dynamic adjustment coefficient is appropriately increased to make full use of the zone capacity; at the same time, the current rainfall data and visibility data are extracted from the real-time meteorological data obtained in step S1. When the rainfall data exceeds the light rain threshold or the visibility data is lower than the good visibility standard, the dynamic adjustment coefficient is further reduced.

[0052] The preferred formula for calculating the dynamic adjustment coefficient is as follows: ; in, For dynamic adjustment coefficients, As the reference adjustment parameter, The growth rate impact coefficient. For the growth rate of tourist numbers, The meteorological impact coefficient. This is the current rainfall. For the maximum rainfall, This is the trend adjustment coefficient. For the trend of changing tourist numbers, This represents the average number of tourists. It is the arctangent function.

[0053] The formula for calculating the current carrying capacity threshold is as follows: ; in, This is the current carrying threshold. For movable area, This represents the standard value for per capita area. For safe evacuation capacity values, For dynamic adjustment coefficients, This is the actual temperature. For the optimal temperature, This refers to the temperature adjustment range.

[0054] It should be noted that the current carrying capacity threshold changes dynamically with the predicted trend of tourist flow and real-time weather conditions, which is more forward-looking and adaptable than a fixed benchmark carrying capacity threshold; the value range of the dynamic adjustment coefficient is usually within a certain range to avoid excessive threshold adjustment leading to loss of management control.

[0055] S2.7: Calculate the ratio of the predicted tourist flow to the current carrying capacity threshold for the corresponding time period to obtain the carrying capacity pressure index. When the carrying capacity pressure index is greater than the warning trigger ratio threshold, mark this partition as a warning partition and generate a warning partition identifier list.

[0056] Specifically, the predicted tourist number sequence of each zone is traversed. For each predicted time period of each zone, the average predicted tourist number in the predicted tourist flow value of that time period is extracted. The current carrying capacity threshold corresponding to that time period is read, and a division operation is performed to obtain the carrying capacity pressure index. The carrying capacity pressure index reflects the degree of saturation of the predicted tourist volume relative to the dynamic carrying capacity.

[0057] Preferably, the formula for calculating the bearing pressure index is as follows: ; in, The pressure bearing index, Forecast values ​​of tourist traffic This is the current carrying threshold. This is the trend weighting coefficient. For the current number of tourists, This is the number of tourists in the same period in history. It is a symbolic function.

[0058] It should be noted that the bearing pressure index is a dimensionless ratio. When the bearing pressure index is less than one, it indicates that the zoning has sufficient bearing capacity. When the bearing pressure index is close to or exceeds one, it indicates that the zoning is about to reach or exceed its bearing limit.

[0059] Preferably, a warning trigger ratio threshold is set, the load pressure index of all partitions is traversed, and when the load pressure index of a certain partition is greater than the warning trigger ratio threshold in a certain period of time, the partition is marked as a warning partition in that period of time, and a list of warning partition identifiers is generated. Specifically, the setting value of the early warning trigger ratio threshold is obtained from the scenic area operation and management regulations. This threshold is usually set as a small value corresponding to a certain percentage. An early warning judgment loop program is established to check the carrying pressure index of each zone and each time period one by one and perform numerical comparison judgment. When the judgment conditions are met, the corresponding zone identifier, predicted time period identifier, carrying pressure index and current carrying threshold are recorded. This information is combined to form an early warning record item and added to the early warning zone identifier list.

[0060] Preferably, a warning level classification is added to the warning zone identifier list, and the warning level is divided according to the numerical range of the load pressure index; when the load pressure index is in the first interval, it is marked as a level one warning; when the load pressure index is in the second interval, it is marked as a level two warning and the lower limit of the second interval is greater than the upper limit of the first interval; when the load pressure index exceeds the third critical value, it is marked as a level three warning and the third critical value is greater than the upper limit of the second interval.

[0061] It should be noted that the warning zone identification list contains information on all zones identified as having an overload risk and their time periods, providing a target list for subsequent diversion decisions; the warning level classification is used to distinguish different levels of carrying capacity pressure and guide the adoption of differentiated control measures.

[0062] In an optional embodiment, when the carrying capacity pressure index of multiple adjacent zones exceeds the warning trigger ratio threshold for consecutive time periods, it is determined to be a regional congestion risk. This zone is then merged into a warning area cluster, and the overall warning level is upgraded. If the carrying capacity pressure index of a certain zone continues to rise for multiple consecutive time periods without showing a downward trend, a trend warning indicator is added to the warning record of this zone, and priority is given to scheduling and diverting resources. If the number of warning zones in the warning zone indicator list exceeds the preset ratio threshold of the total number of zones in the scenic area, the entire scenic area traffic control plan is activated, and flow restriction measures are implemented at the main entrance of the scenic area. When the carrying capacity pressure index of multiple adjacent zones does not exceed the warning trigger ratio threshold for consecutive time periods, it is determined that the local carrying capacity pressure is controllable, and no regional warning upgrade is triggered, maintaining the existing zone-level warning and diversion mechanism.

[0063] S3: Construct a tourist flow transfer matrix between the early warning zone and adjacent zones, calculate the tourist flow transfer probability value, and obtain the diversion target zone; Specifically, a diversion path planning module and a target partition selection module are established in the scenic area's data center server. The diversion path planning module is configured with a geographic topology database and a path calculation engine. The target partition selection module is configured with a weight evaluator and a probability sorter. The two modules exchange information through a data interface. The diversion path planning module transmits spatial relationship data to the target partition selection module to generate diversion schemes.

[0064] S3.1: Extract the list of warning zone identifiers, query the scenic area's geographic topology database, identify all adjacent zones that have direct connections with the warning zones, and construct a set of adjacent zones; Specifically, the boundary geometry information and passage connection data of each zone of the scenic area are read from the geographic topology database; the passage connection data records whether there are walkways, boardwalks or cable car lines that tourists can pass between any two zones; each warning zone in the warning zone identifier list is traversed, and the zone identifiers of other zones that are connected to the warning zone are retrieved, and the zone identifiers of these zones are summarized to form the corresponding set of adjacent zones.

[0065] It should be noted that the adjacent partition set is indexed by the partition identifier of the warning partition and contains a list of all partitions that are physically connected to the warning partition; direct passage connection refers to adjacent partitions that tourists can reach from the warning partition without passing through a third partition; the adjacent partition set provides a candidate range for subsequent diversion target selection.

[0066] For example, area A, the core viewing platform of a scenic spot, is marked as a warning zone. By querying the geographic topology database, it is found that area A is connected to area B, the ecological forest area, via the east-side walkway; to area C, the leisure plaza area, via the west-side boardwalk; and to area D, the mountaintop sightseeing area, via the south-side cable car line. However, there is no direct passage between area A and area E, the parking lot area, to the north. Therefore, the set of adjacent zones of area A includes the zone identifiers of areas B, C, and D, but does not include area E.

[0067] S3.2: Based on the coordinate system of the scenic area digital map, extract the coordinates of the center point of each zone in the early warning zone and the adjacent zone set, calculate the spatial straight-line distance, and obtain the inter-zone distance parameters; Specifically, the vertex coordinate sequence of the boundary polygons of each zone is read from the geographic topology database, and the centroid calculation algorithm is used to calculate the arithmetic mean of the vertex coordinates of the polygons to obtain the center point coordinates; the center point coordinates include the x-coordinate component and the y-coordinate component; for each adjacent zone in the set of adjacent zones of the warning zone, the center point coordinates of the two are extracted, and the spatial straight-line distance between the two points is calculated by applying the Euclidean distance formula.

[0068] Preferably, the formula for calculating the interval distance parameter is as follows: ; in, This is the distance parameter between partitions p and q. Weights for coordinate dimensions Let p be the coordinates of the partition in the k-th dimension. Let q be the coordinates of the partition in the k-th dimension. This is the terrain correction factor. The elevation difference between partitions p and q. This represents the average elevation difference.

[0069] It should be noted that the distance parameter between zones reflects the spatial scale that tourists need to traverse from the warning zone to the adjacent zone. The shorter the distance, the more convenient the transfer. Although the straight-line distance is not the same as the actual walking path length, it can effectively characterize the spatial proximity between zones.

[0070] S3.3: The distance parameters of the intervals are normalized using the distance decay function, and the distance weights from the warning partition to each adjacent partition in the adjacent partition set are calculated. Specifically, the distance parameters between the warning zone and all adjacent zones in the adjacent zone set are calculated, and the maximum and minimum distance values ​​are obtained. An exponentially decaying distance weight function is constructed. The input of the distance weight function is the distance parameter between the zones, and the output is a weight value between zero and one. When the distance parameter between the zones is equal to the minimum distance value, the distance weight close to one indicates a high tendency to transfer. When the distance parameter between the zones is equal to the maximum distance value, the distance weight close to zero indicates a low tendency to transfer.

[0071] S3.4: Obtain the attraction type identifiers, visitor satisfaction rating data, and historical visit popularity data for each zone from the scenic area management information system, and comprehensively calculate the landscape attractiveness index of each zone; Specifically, the attraction type identification will classify the areas into natural landscape type, cultural landscape type, amusement facility type and service support type; the tourist satisfaction score data of each area will be extracted from the tourist evaluation database. The tourist satisfaction score data is the comprehensive evaluation score of tourists on the landscape quality, facility conditions and service level of each area; the number of tourist visits to each area in the past time period will be counted from the historical operation data, and the proportion of the visit frequency to the total number of visits to the whole scenic area will be calculated as historical visit popularity data.

[0072] The preferred formula for calculating the landscape attractiveness index is as follows: ; in, As a landscape attractiveness index, Rate tourist satisfaction. This represents the coefficient for the type of tourist attraction. Based on historical visit popularity, This is the heat adjustment coefficient. For the age parameters of the scenic spot, Rate the facilities' completeness. Rate the facility with the highest rating.

[0073] It should be noted that the landscape attractiveness index comprehensively reflects the attractiveness of a zone to tourists. The higher the index value, the stronger the zone's attraction to tourists and its potential to divert and accommodate visitors. Different attraction types correspond to different type scores, with natural landscape types and cultural landscape types typically scoring higher than service-related types.

[0074] For example, Area B of the ecological forest area is a natural landscape type, and the average score of the visitor satisfaction rating data is a high value. The historical visit popularity data shows that the daily average number of visitors accounts for a certain percentage of the total number of visitors in the scenic area. Area C of the leisure square is a service support type, with high visitor satisfaction but relatively low historical visit popularity. Through weighted calculation, the landscape attractiveness index of Area B is higher than that of Area C, indicating that Area B is more suitable as a diversion target.

[0075] S3.5: Based on the landscape attractiveness index, the normalization method is used to calculate the landscape attractiveness weight of each adjacent partition in the set of adjacent partitions from the warning partition; Specifically, the landscape attractiveness index of all adjacent partitions in the adjacent partition set is extracted, and the sum of these index values ​​is calculated; the landscape attractiveness index of each adjacent partition is divided by the sum to obtain the landscape attractiveness weight of that adjacent partition; the landscape attractiveness weight represents the relative strength of the attraction of that partition to tourists among all candidate adjacent partitions.

[0076] S3.6: A weighted fusion strategy is adopted to linearly combine the distance weight and the landscape attraction weight to calculate the probability of tourist flow transfer from the warning zone to each adjacent zone in the adjacent zone set; Specifically, a distance factor contribution ratio coefficient and an attraction factor contribution ratio coefficient are set, and the sum of the two coefficients equals one. For the pairing relationship between the warning zone and each adjacent zone, the corresponding distance weight and landscape attraction weight are extracted, and the weighted sum is obtained according to the distance factor contribution ratio coefficient and the attraction factor contribution ratio coefficient to obtain the probability of tourist flow transfer.

[0077] The preferred formula for calculating the probability of tourist flow shifting is as follows: ; in, For the probability of tourist flow shift, Distance weights As a weight for landscape attractiveness, The distance factor contributes a proportionality coefficient, where N is the number of candidate partitions. This is the capacity adjustment coefficient. This represents the remaining carrying capacity value. This is the capacity threshold.

[0078] It should be noted that the probability of tourist flow transfer comprehensively considers two core factors: spatial distance convenience and landscape attractiveness, and quantitatively describes the likelihood of tourists transferring from the warning zone to adjacent zones; the contribution ratio coefficients of distance factor and attractiveness factor can be dynamically adjusted according to the characteristics of the scenic area and management strategy.

[0079] S3.7: Organize the probability of tourist flow transfer between the warning zone and all adjacent zones in the adjacent zone set in matrix form to construct a tourist flow transfer matrix; Specifically, a two-dimensional matrix data structure is established, where the row index of the matrix corresponds to each warning partition in the warning partition identifier list, and the column index of the matrix corresponds to all possible adjacent partitions; the matrix is ​​filled with the corresponding tourist flow transfer probability values, and for partition pairs that do not have a direct channel connection, the matrix element value is set to zero; each row of the tourist flow transfer matrix is ​​normalized so that the sum of the transfer probabilities from each warning partition to all its adjacent partitions is equal to one.

[0080] Preferably, the tourist flow transfer matrix is ​​denoted in matrix symbol form, where the matrix element values ​​represent the transfer probability from the warning zone to the adjacent zone, and the sum of the rows of the matrix equals one to ensure the completeness of the probability distribution.

[0081] It should be noted that the tourist flow transfer matrix is ​​a sparse matrix, storing non-zero probability values ​​only for locations in partitions with channel connections; the tourist flow transfer matrix provides a quantitative data basis for tourist flow tendency for diversion decisions.

[0082] S3.8: Extract the current carrying capacity threshold and real-time tourist flow data of each zone, calculate the remaining carrying capacity value of each zone, and filter the zones with remaining carrying capacity values ​​greater than zero as a candidate diversion zone set; Specifically, it iterates through all zones of the scenic area, reads the current carrying capacity threshold of each zone in the current time period, obtains the current number of tourists in the zone from the visitor flow statistics device as real-time tourist flow data, performs a subtraction operation, subtracts the real-time tourist flow data from the current carrying capacity threshold to obtain the remaining carrying capacity value, and determines whether the remaining carrying capacity value is greater than zero, adding the zone identifier of the zone that meets the condition to the candidate diversion zone set.

[0083] Preferably, the formula for calculating the remaining carrying capacity is as follows: ; in, This represents the remaining carrying capacity value. This is the current carrying threshold. For the current number of tourists, This is the trend prediction coefficient. The rate of change in tourist numbers, For predicting time intervals, Meteorological influencing factors It should be noted that the remaining capacity value represents the number of additional visitors that a zone can currently accommodate; only zones with a remaining capacity value greater than zero are eligible to accept diverted visitors, and the candidate diversion zone set excludes zones that have reached their capacity limit.

[0084] S3.9: For each warning zone, extract the tourist flow transfer probability from the tourist flow transfer matrix to each of the neighboring zones in the neighboring zone set, filter the neighboring zones that are simultaneously in the candidate diversion zone set, and select the zone with the highest tourist flow transfer probability among the neighboring zones that meet the conditions, and mark it as the diversion target zone. Specifically, each warning partition in the warning partition identifier list is traversed, and the probability value of the corresponding row in the tourist flow transfer matrix is ​​read; the intersection operation of the set of adjacent partitions of this warning partition and the set of candidate diversion partitions is performed to obtain the effective diversion candidate partitions that are both adjacent partitions and have remaining capacity; the tourist flow transfer probability values ​​corresponding to the effective diversion candidate partitions are compared, and the maximum tourist flow transfer probability is identified; the partition corresponding to the maximum tourist flow transfer probability is taken as the diversion target partition of this warning partition.

[0085] Preferably, a mapping table is established between the warning zones and the diversion target zones, recording the zone identifier of the warning zone, the zone identifier of the corresponding diversion target zone, the maximum probability of tourist flow transfer, the remaining carrying capacity of the diversion target zone, and the recommended number of tourists to be diverted.

[0086] It should be noted that the diversion target zone is the optimal diversion direction after comprehensively considering the convenience of distance, landscape attractiveness and carrying capacity margin; when all adjacent zones of a certain warning zone are not in the candidate diversion zone set, it indicates that there are no available diversion targets around the warning zone, and it is necessary to start traffic diversion measures on a larger scale.

[0087] In an optional embodiment, when the remaining carrying capacity of the diversion target zone is less than the expected number of diverted tourists in the warning zone, multiple diversion target zones are selected in descending order of tourist flow transfer probability for combined diversion. The number of tourists allocated to each diversion target zone is allocated according to the tourist flow transfer probability ratio and the remaining carrying capacity. When the connection channel between the warning zone and the diversion target zone itself has a bottleneck in its capacity, a channel flow restriction indicator is added to the mapping table to avoid channel congestion during the diversion process. When the same zone is repeatedly selected as the diversion target zone in multiple consecutive time periods, the landscape attractiveness weight of that zone is dynamically reduced to guide tourists to disperse to more diverse zones.

[0088] S4: Push guidance information for the diversion target zone to the visitor terminal of the warning zone, and simultaneously adjust the entrance gate of the warning zone to reduce the passage speed and increase the passage speed of the entrance gate of the diversion target zone.

[0089] Specifically, an intelligent push service module and a gate control and scheduling module are established in the scenic area's data center server. The intelligent push service module is configured with a visitor terminal communication interface and an information generation engine. The gate control and scheduling module is configured with a gate device driver and a passage rate calculator. The two modules are synchronously triggered through a real-time control bus. While sending push instructions, the intelligent push service module sends a linkage control signal to the gate control and scheduling module. After receiving the linkage control signal, the gate control and scheduling module performs gate parameter adjustment operations.

[0090] S4.1: Extract the list of tourist identity identifiers currently located in the warning zone from the scenic area tourist management system, and obtain the mobile terminal device identifier and communication address of each tourist by associating with the tourist registration information; Specifically, when collecting real-time visitor flow data, the visitor flow statistics equipment simultaneously records the identity information of visitors entering each zone. The visitor identity information is obtained through the scenic area ticket QR code, facial recognition device, or electronic wristband. The system queries the scenic area's visitor registration database to extract the mobile phone number, mobile application user identifier, or WeChat mini-program open identifier corresponding to the visitor's identity as the mobile terminal device identifier. It also obtains the communication addresses used for message push, including SMS channel address, application push channel address, and mini-program push channel address.

[0091] It should be noted that the tourist identification list uses the zone identifier as the group index to record all tourists currently staying in the warning zone; the mobile terminal device identifier is used to accurately locate the push target, and the communication address provides the technical path for message transmission.

[0092] For example, the core observation deck area A is marked as a warning zone. The system query finds that there are a certain number of tourists currently staying in area A. Among them, a certain number of tourists entered the scenic area through the official mobile application and authorized their location information, and a certain number of tourists purchased tickets and bound their accounts through the WeChat mini program. The system extracts the mobile terminal device identifiers of these tourists, including the user's unique identifier in the application and the WeChat OpenID; at the same time, it obtains the mobile phone numbers of tourists who have authorized SMS push as backup communication addresses.

[0093] S4.2: Extract the target diversion zone and the corresponding tourist flow transfer probability, and read the scenic spot graphic description data, tour route navigation data and current activity discount data of the target diversion zone from the scenic area content management system; Specifically, the system reads the partition identifier of the diversion target partition corresponding to the warning partition from the mapping table, and extracts the tourist flow transfer probability value; it accesses the multimedia resource library of the scenic area content management system to obtain the scenic spot image and text description data of the diversion target partition, which includes the scenic spot name, feature description, high-definition photos and video materials; it calls the scenic area navigation service interface to generate tour route navigation data from the warning partition to the diversion target partition, which includes the route distance, estimated walking time and key nodes passed through; and it queries the scenic area operation activity database to extract the current activity discount data of the diversion target partition, including catering discount information, souvenir coupons and free places for special experience projects.

[0094] Preferably, the display priority of the scenic spot graphic and textual description data is sorted according to the numerical value of the tourist flow transfer probability. The higher the probability value, the higher the display position of the description content in the push information of the diversion target area.

[0095] It should be noted that the data on scenic spot images and text is used to enhance the attractiveness of the diversion target areas to tourists, the tour route navigation data reduces the cognitive cost for tourists to reach the diversion target areas, and the current activity discount data incentivizes tourists to respond to the diversion guidance through actual benefits.

[0096] S4.3: Based on tourists' historical travel preferences and age group identifiers, a content matching algorithm is used to filter content fragments in the scenic spot's text and image descriptions that match tourists' interests, and generate personalized guidance text; Specifically, historical visit preference data is extracted from visitor behavior records, including the types of areas visited, the distribution of stay duration, and consumption behavior characteristics. Age group identifiers are calculated based on the birth dates provided by visitors during registration, categorizing them into teenagers, young adults, middle-aged individuals, and seniors. A preference-content matching rule base is established, defining content recommendation strategies corresponding to different preference types and age groups. For visitors who prefer natural landscapes, ecological environment descriptions and information on flora and fauna are extracted from scenic spot illustrations. For visitors who prefer cultural landscapes, historical and cultural background information and architectural features are extracted. For seniors, personalized guidance text includes descriptions of accessible facilities and rest area locations.

[0097] Preferably, the personalized guide text generation formula adopts a template filling method. The template structure includes a greeting, a description of the current congestion status of the current section, reasons for recommending the target section, a display of tour path navigation data, and a presentation of current activity discount data.

[0098] It should be noted that personalized guidance text enhances the appeal and persuasiveness of push notifications to different tourists through content customization, resulting in a higher conversion response rate compared to uniform mass messages; the depth of analysis of historical travel preference data affects the accuracy of content matching.

[0099] S4.4: Integrate personalized guidance text, tour route navigation data, and current activity discount data into a structured push message package, and send the push message package to the mobile terminal devices of tourists in the warning zone according to the mobile terminal device identifier and communication address through the message push interface of the scenic area's mobile application and the push interface of WeChat mini program; Specifically, the data structure of the push message package is constructed, including message type identifier, priority level, personalized guidance text field, tour route navigation data field, current activity discount data field, and interactive response interface; the priority level is set to high priority, and the push message package is displayed at the top of the notification bar of the tourist's mobile terminal device; the application programming interface of the mobile push service platform is called, and the content parameters of the push message package, the list of mobile terminal device identifiers, and the push timestamp are used as interface input parameters; the push service platform matches the corresponding push channel according to the mobile terminal device identifier, and transmits the push message package to the tourist's mobile terminal device through a long connection channel or SMS gateway.

[0100] Preferably, a one-click navigation button and a coupon redemption button are embedded in the push message package. The one-click navigation button is associated with the scenic area map application and automatically loads the tour route navigation data. After the coupon redemption button is triggered, the electronic coupon in the current activity discount data is stored in the tourist's account.

[0101] It should be noted that the push message package uses a rich media format to support mixed text and images and interactive operations, improving the intuitiveness and convenience of information delivery; the frequency of push interface calls is limited by the platform, and repeated pushes to the same visitor in a short period of time should be avoided to prevent disturbance.

[0102] In an optional embodiment, the delivery status of push message packages and the reading status of tourists are monitored, and the mobile terminal devices of tourists that have not been delivered or read are recorded in the resend queue; when a tourist's mobile terminal device is offline, the push message package is automatically resent after the device is back online; when a tourist clicks the one-click navigation button in the push message package, the tourist's response status is marked on the scenic area operation monitoring screen, and the response rate index of diversion guidance is calculated; if a tourist has not left the warning zone within a certain time interval after the push message package is sent, a second reminder push is sent, which adds a congestion warning and alternative diversion route suggestions.

[0103] S4.5: Extract the load-bearing pressure index of the warning zone, and calculate the passage speed reduction coefficient of the entrance gate of the warning zone based on the load-bearing pressure index; Specifically, the load-bearing pressure index value corresponding to the warning zone is read from the warning zone identifier list; a traffic rate control function is established, which takes the load-bearing pressure index as the independent variable and outputs a traffic rate reduction ratio coefficient; when the load-bearing pressure index is in the first-level warning zone, the traffic rate reduction ratio coefficient is set to a small value for mild traffic restriction; when the load-bearing pressure index is in the second-level warning zone, the traffic rate reduction ratio coefficient is increased to a medium value for moderate traffic restriction; when the load-bearing pressure index is in the third-level warning zone, the traffic rate reduction ratio coefficient is increased to a large value for severe traffic restriction.

[0104] It should be noted that the passage rate reduction ratio quantifies the reduction in the frequency of gate opening; the larger the value, the stronger the flow restriction. The gate control parameters are automatically adjusted by driving the load pressure index, avoiding delays and subjective biases caused by manual judgment.

[0105] S4.6: Obtain the current passage rate parameter of the entrance gate of the warning zone, multiply the current passage rate parameter by the difference of the passage rate reduction ratio coefficient, and calculate the target passage rate parameter of the entrance gate of the warning zone. Specifically, the current passage rate parameter is read from the gate equipment controller of the warning zone. The current passage rate parameter represents the number of visitors allowed to pass through the gate per unit time. The passage rate reduction ratio coefficient is subtracted from one to obtain the retention ratio coefficient. The current passage rate parameter is multiplied by the retention ratio coefficient to obtain the target passage rate parameter.

[0106] Preferably, the formula for calculating the target traffic speed parameter is as follows: ; in, For the target traffic rate parameter, This is the current traffic rate parameter. To reduce the scaling factor for traffic speed, The neighborhood influence coefficient. For adjacent partition capacity, This represents the current partition capacity. For average partition size, For ideal traffic speed, This represents the current traffic flow rate.

[0107] It should be noted that the target passage rate parameter is lower than the current passage rate parameter. By reducing the passage efficiency of the turnstiles, the speed at which tourists enter the warning zone is slowed down, thereby controlling the inflow of tourists from the supply side. The target passage rate parameter must not be lower than the minimum safe operating rate of the turnstile equipment to avoid excessive flow restriction causing congestion at the entrance.

[0108] S4.7: Calculate the gate opening duration and gate closing duration of the early warning zone entrance gate based on the target passage rate parameters, generate gate control commands and send them to the control unit of the early warning zone entrance gate; Specifically, the passage rate of the turnstile is determined by the single passage cycle, which is equal to the sum of the gate opening time and the gate closing time. The single passage cycle value is calculated based on the reciprocal of the target passage rate parameter. Within a single passage cycle, the gate opening time is allocated for visitor passage, and the gate closing time is allocated for gate reset and safety intervals. Turnstile control commands containing gate opening and closing time parameters are encapsulated and transmitted to the control unit of the warning zone entrance turnstile via an industrial bus protocol or wireless control protocol.

[0109] Preferably, the formulas for calculating the gate opening time and gate closing time are as follows: ; in, For the duration the gate is open, The duration of the gate closure. For safe passage distance, Walking speed, For reaction time, For the target traffic rate parameter, To enable the adjustment coefficient, Minimum shutdown time, For safety interval factor, For population density, This represents the maximum population density.

[0110] It should be noted that after the gate control command is issued, the control unit immediately adjusts the gate motor drive parameters and controls the gate opening and closing actions according to the new time parameters; the gate opening time must ensure the safe passage of tourists at a normal walking speed, and the gate closing time includes mechanical reset time and anti-tailgating detection time.

[0111] S4.8: Extract the remaining carrying capacity value of the diversion target zone and calculate the passage speed increase ratio coefficient of the entrance gate of the diversion target zone; Specifically, the remaining capacity value of the diversion target zone is read from the mapping table. The remaining capacity value represents the additional number of tourists that the diversion target zone can currently accommodate. The baseline capacity threshold of the diversion target zone is obtained, and the ratio of the remaining capacity value to the baseline capacity threshold is calculated as the capacity adequacy index. The passage speed increase ratio coefficient is set according to the capacity adequacy index. The higher the capacity adequacy index, the stronger the acceptance capacity of the diversion target zone, and the larger the corresponding passage speed increase ratio coefficient.

[0112] It should be noted that the passage speed increase ratio takes into account the carrying capacity and attractiveness level of the diversion target area. When the capacity is sufficient and the attractiveness is strong, the passage efficiency of the gate is increased to accelerate the flow of tourists to the diversion target area. The passage speed increase ratio has an upper limit threshold to avoid the gate from running too fast and affecting safety.

[0113] S4.9: Obtain the current passage rate parameter of the entrance gate of the diversion target zone, add the current passage rate parameter to the passage rate increase ratio coefficient, calculate the target passage rate parameter of the entrance gate of the diversion target zone, generate the corresponding gate control command and send it to the entrance gate of the diversion target zone. Specifically, the current passage rate parameter is read from the gate device controller of the diversion target zone; the current passage rate parameter is added to the product of the current passage rate parameter and the passage rate increase ratio coefficient to obtain the target passage rate parameter of the entrance gate of the diversion target zone; following the same method as in step S4.7, the gate opening time and gate closing time are calculated based on the target passage rate parameter, and the shortening of the single passage cycle increases the number of people passing through per unit time; the gate control command is encapsulated and sent to the entrance gate of the diversion target zone through the control network.

[0114] Preferably, the calculation formula for the target passage rate parameter of the diversion target zone entrance gate is as follows: ; in, The target throughput rate parameters for the entrance gates of the diversion target zones. This is the current traffic rate parameter. To increase the rate of increase in traffic flow, To optimize the adjustment coefficient, For optimal passage speed, For the range of traffic speeds, For the probability of tourist flow shift, This represents the maximum transition probability.

[0115] It should be noted that the target passage rate parameter of the entrance gate of the diversion target zone is higher than the current passage rate parameter. By speeding up the passage speed of the gate, tourists are encouraged to enter the diversion target zone, thereby enhancing the attractiveness of diversion from the demand side. The increased target passage rate parameter still needs to meet the maximum operating speed limit and safety specifications of the gate equipment.

[0116] In an optional embodiment, a real-time monitoring mechanism for diversion effect is established to continuously collect real-time visitor flow data of the warning zone and the diversion target zone, and calculate the visitor flow change rate before and after diversion. When the real-time visitor flow data of the warning zone shows a downward trend within a preset time window after gate control, and the real-time visitor flow data of the diversion target zone shows an upward trend, the diversion measures are deemed effective, and the current gate control parameters are maintained. When the carrying pressure index of the warning zone continues to rise without relief, the target passage rate parameter of the entrance gate of the warning zone is further reduced, or a second diversion target zone is added and guidance information is pushed to it. When the real-time visitor flow data of the diversion target zone approaches its current carrying threshold, the diversion of visitors to that zone is stopped, the push message packet is switched to recommend other diversion target zones, and the target passage rate parameter of the entrance gate of that zone is restored to the normal level. The diversion response rate index and carrying balance index are calculated periodically to evaluate the attractiveness of personalized guidance text and the effectiveness of gate rate control. Based on the evaluation results, the calculation parameters of the passage rate reduction ratio coefficient and the passage rate increase ratio coefficient are optimized.

[0117] In summary, this invention constructs a tourist flow prediction model based on LSTM neural networks, incorporating historical data, holiday factors, and meteorological parameters. This provides a 30-minute to 2-hour advance prediction window for management decisions, establishing a time advantage from the outset. By dynamically adjusting the carrying capacity threshold and introducing a carrying capacity pressure index, it overcomes the limitations of traditional fixed thresholds, achieving a more precise and adaptive early warning mechanism and effectively avoiding false alarms and missed alarms. By utilizing a tourist flow transfer matrix to comprehensively quantify distance weights and landscape attractiveness weights, it transforms the selection of diversion targets from experience-based judgment to data-driven scientific decision-making, significantly improving the success rate of diversion schemes and tourist acceptance. Through a combination of information push and gate speed adjustment, it achieves a dual synergy of flexible guidance and rigid control, ensuring rapid response and precise execution of diversion measures. This maximizes the scenic area's carrying capacity and service quality while ensuring tourist safety.

[0118] This embodiment also provides a computer device applicable to the smart scenic area zoning and flow control method based on tourist flow prediction and dynamic carrying capacity control, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the smart scenic area zoning and flow control method based on tourist flow prediction and dynamic carrying capacity control proposed in the above embodiment.

[0119] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0120] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the smart scenic area zoning and flow control method based on tourist flow prediction and dynamic carrying capacity control as proposed in the above embodiments.

[0121] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0122] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A smart scenic area zoned flow control method based on tourist flow prediction and dynamic carrying capacity control, characterized in that: include, Collect historical and real-time visitor flow data for each zone of the scenic area, construct a visitor flow prediction model, and generate predicted visitor flow values ​​for each zone for future periods. The baseline carrying capacity threshold is calculated based on the spatial capacity parameters and safe evacuation parameters of each zone. The current carrying capacity threshold is then dynamically adjusted in conjunction with the predicted tourist flow to obtain the warning zones. Construct a tourist flow transfer matrix between the warning zone and adjacent zones, calculate the tourist flow transfer probability value, and obtain the diversion target zone; The system pushes guidance information for the diversion target zone to the visitor terminal in the warning zone, and simultaneously adjusts the entrance gate of the warning zone to reduce the passage speed and increases the passage speed of the entrance gate of the diversion target zone.

2. The smart scenic area zoned flow control method based on tourist flow prediction and dynamic carrying capacity control as described in claim 1, characterized in that: The method for obtaining the target partition for traffic splitting is as follows: Iterate through each warning zone in the warning zone identifier list and read the probability value of the corresponding row in the tourist flow transfer matrix; Perform an intersection operation between the set of adjacent partitions of this warning partition and the set of candidate diversion partitions to obtain a valid diversion candidate partition that is both an adjacent partition and has remaining capacity. Compare the tourist flow transfer probability values ​​corresponding to the effective diversion candidate zones, and identify the maximum tourist flow transfer probability; The partition corresponding to the maximum probability of tourist flow transfer is used as the diversion target partition for this early warning partition.

3. The smart scenic area zoned flow control method based on tourist flow prediction and dynamic carrying capacity control as described in claim 2, characterized in that: The method for constructing the tourist flow transfer matrix is ​​as follows: Extract the list of warning zone identifiers, query the scenic area's geographic topology database, identify all adjacent zones that have direct connections to the warning zones, and construct a set of adjacent zones; Based on the coordinate system of the scenic area digital map, the center point coordinates of each partition in the early warning partition and the adjacent partition set are extracted, the spatial straight-line distance is calculated, and the inter-partition distance parameter is obtained. The distance parameters of the intervals are normalized using a distance decay function, and the distance weights from the warning partition to each adjacent partition in the adjacent partition set are calculated. The scenic area management information system is used to obtain the attraction type identification, visitor satisfaction rating data and historical visit popularity data for each zone, and the landscape attractiveness index of each zone is calculated in a comprehensive manner. Based on the landscape attractiveness index, the landscape attractiveness weights of each adjacent partition in the adjacent partition set to the warning partition are calculated using a normalization method. A weighted fusion strategy is used to linearly combine the distance weight and the landscape attraction weight to calculate the probability of tourist flow transfer from the warning zone to each adjacent zone in the adjacent zone set; The probability of tourist flow transfer between the warning zone and all adjacent zones in the adjacent zone set is organized in matrix form to construct a tourist flow transfer matrix.

4. The smart scenic area zoned flow control method based on tourist flow prediction and dynamic carrying capacity control as described in claim 3, characterized in that: The method for generating the early warning zone identifier list is as follows: Extract the predicted tourist flow values ​​to obtain the predicted tourist number sequence for each zone in consecutive future time periods; Calculate the dynamic adjustment coefficient based on the numerical change trend in the predicted tourist number sequence and real-time meteorological data; Multiply the baseline load threshold by the dynamic adjustment coefficient to obtain the current load threshold of each partition in the corresponding time period; The ratio of the predicted tourist flow to the current carrying capacity threshold for the corresponding time period is calculated to obtain the carrying capacity pressure index. When the carrying capacity pressure index is greater than the warning trigger ratio threshold, this partition is marked as a warning partition, and a warning partition identifier list is generated.

5. The smart scenic area zoned flow control method based on tourist flow prediction and dynamic carrying capacity control as described in claim 4, characterized in that: Spatiotemporal correlation analysis is performed on the warning zones in the warning zone identifier list to detect whether there are adjacent zones that issue warnings simultaneously in similar time periods; When the carrying pressure index of multiple adjacent partitions exceeds the warning trigger ratio threshold in a continuous period of time, it is determined to be a regional congestion risk, this partition is merged into a warning area cluster, and the overall warning level is upgraded. If the load-bearing pressure index of a certain partition continues to rise for several consecutive periods without showing a downward trend, a trend warning indicator will be added to the warning record of this partition, and resources will be prioritized for allocation and diversion. If the number of warning zones in the warning zone identifier list exceeds the preset ratio threshold of the total number of zones in the scenic area, the entire scenic area traffic control plan will be activated, and traffic restriction measures will be implemented at the main entrance of the scenic area. When the load-bearing pressure index of multiple adjacent zones does not exceed the warning trigger ratio threshold within a continuous period, it is determined that the local load-bearing pressure is controllable, and the regional warning upgrade is not triggered, maintaining the existing zone-level warning and diversion mechanism.

6. The smart scenic area zoned flow control method based on tourist flow prediction and dynamic carrying capacity control as described in claim 4, characterized in that: The method for obtaining the benchmark bearing threshold is as follows: Obtain topographic mapping data for each zone of the scenic area, extract the activity area, main passage width, and number of entrances and exits for each zone, and construct a set of spatial capacity parameters. Based on the terrain features and tour methods of each zone, a standard value for the area occupied per person is set. The spatial carrying capacity of each zone is calculated by dividing the value of the usable area by the standard value of the area occupied per person. Obtain the total length data of evacuation routes and emergency evacuation time limits for each zone, and calculate the safe evacuation capacity of each zone based on the number of entrances and exits, the width of the main passage, and the total length data of the evacuation routes. The spatial carrying capacity value and the safe evacuation capacity value are compared, and the smaller one is selected as the benchmark carrying capacity threshold for each zone.

7. The smart scenic area zoned flow control method based on tourist flow prediction and dynamic carrying capacity control as described in claim 4, characterized in that: The method for obtaining the predicted tourist flow value is as follows: Obtain holiday date information for the area where the scenic spot is located, calculate the holiday impact weight coefficient based on the holiday type and duration, and construct a holiday factor vector; Real-time meteorological data is obtained by collecting current temperature, humidity, rainfall, and visibility data of the scenic area through meteorological monitoring equipment. Statistical regression analysis was performed on the meteorological data and tourist flow values ​​for the corresponding dates in the historical tourist flow data to calculate the impact coefficient of different meteorological conditions on tourist flow and obtain weather correction parameters. The historical tourist flow data is divided into a training dataset and a validation dataset according to time order, and the temporal and spatial feature variables are extracted from the training dataset. A tourist flow prediction model based on a long short-term memory neural network is constructed. The training sample matrix is ​​input into the tourist flow prediction model for training. The model parameters are iteratively optimized using the validation dataset. The real-time tourist flow data, the holiday factor vector corresponding to the current date, and the real-time weather data are input into the trained tourist flow prediction model to generate a predicted tourist flow value.

8. The smart scenic area zoned flow control method based on tourist flow prediction and dynamic carrying capacity control as described in claim 7, characterized in that: The specific formula for the predicted tourist flow is as follows: ； in, This represents the predicted tourist flow for zone p in time period t. It is the Sigmoid activation function. Historical traffic weighting coefficient Let be the LSTM hidden states for the first i time periods. This represents the impact coefficient of holidays. Single-hot encoding for holiday types, This is a meteorological correction factor. For meteorological parameter vectors, The current temperature. For optimal temperature during visit, For rainfall, As a temperature regulation factor, It is a rainfall regulating factor.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the smart scenic area zoning and flow control method based on tourist flow prediction and dynamic carrying capacity control as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the smart scenic area zoning and flow control method based on tourist flow prediction and dynamic carrying capacity control as described in any one of claims 1 to 8.

Images