Smart park safety and efficiency optimization system and method based on Internet of Things and AI

Abstract

The invention relates to the technical field of logistics park management, in particular to a smart park safety and efficiency optimization system and method based on Internet of Things and AI, and the system comprises a vehicle access park module, a platform scheduling optimization module, a visitor management module, a perimeter safety module, an AI recognition analysis module, and a park energy consumption management module. According to the invention, the parking efficiency is improved and the waiting time is shortened by analyzing the information of vehicles and cargoes entering the park and dispatching platform allocation, and the cargo loading and unloading position is calculated by using the real-time vehicle position information, so that the residence time of the cargoes in the transfer process is shortened; according to the system, the behaviors of workers are monitored in real time, dangerous behaviors and incorrect equipment wearing personnel are recognized, the accident response speed and the safety management level are improved, the energy utilization rate is increased through facilities such as water and electricity energy consumption collection, and intelligent and green park construction is further promoted.

Description

Technical Field

[0001] This invention relates to the field of logistics park management technology, and in particular to a smart park safety and efficiency optimization system and method based on the Internet of Things and AI. Background Technology

[0002] The field of logistics park management technology focuses on improving operational efficiency and safety within logistics parks, including facility management, warehousing optimization, supply chain coordination and resource scheduling. It utilizes information technology, the Internet of Things, artificial intelligence, and big data analytics to achieve real-time data monitoring, automated operation, and intelligent decision support. This includes vehicle dispatch management, inventory control, energy consumption management, and safety protection measures, aiming to reduce human error, improve response speed, lower operating costs, and enhance economic efficiency and safety standards.

[0003] The Logistics Park Safety and Efficiency Optimization System aims to improve the safety management and operational efficiency of logistics parks. By deploying various intelligent sensors and devices, including video surveillance and environmental monitors, and applying multiple algorithms for data analysis and processing, it monitors logistics activities in real time, automatically identifies and responds to safety risks and efficiency bottlenecks, optimizes cargo flow, improves energy efficiency, enhances safety early warning and accident response capabilities, and provides an automated and intelligent operating environment for logistics parks, thereby improving operational quality and safety levels.

[0004] Traditional logistics park management methods rely on manual vehicle dispatching and cargo loading and unloading, resulting in low efficiency and susceptibility to human error, increasing resource waste and operating costs. They depend on regular inspections and post-incident handling, which are inadequate in responding to emergencies and making it difficult to effectively prevent accidents. They cannot adjust air conditioning and lighting systems according to real-time environmental changes, leading to energy waste and poor environmental conditions, which affect the work efficiency and health of staff. Furthermore, they cannot adjust air conditioning and lighting parameters according to real-time personnel distribution and activity intensity, resulting in some areas being excessively hot or cold or having insufficient light, which also affects staff work efficiency. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a logistics park safety and efficiency optimization system and method based on the Internet of Things and AI.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a smart park security and efficiency optimization system based on the Internet of Things and AI, comprising:

[0007] The vehicle entry and exit module analyzes and identifies vehicle type, cargo destination and expected stay time based on real-time vehicle entry data, calculates the entry and exit sequence and route configuration of multiple vehicles, and obtains a logistics vehicle scheduling list.

[0008] The platform scheduling optimization module calculates the loading and unloading positions of vehicles based on the logistics vehicle scheduling list, combined with the real-time location of vehicles and the destination of goods, and generates loading and unloading position allocation results.

[0009] Based on the loading and unloading location allocation results, the visitor management module collects real-time traffic flow data within the logistics park, calculates the optimal logistics route, and generates park traffic management parameters.

[0010] Based on the park traffic management parameters, the perimeter security module uses image acquisition equipment to collect and analyze the safety equipment wearing status of staff in real time, and generates compliance monitoring records.

[0011] Based on the compliance monitoring records, the AI ​​recognition and analysis module collects images of workers' operations in real time, identifies workers' safety violations and dangerous behaviors, sends early warning information to managers, and generates risk behavior response results.

[0012] Based on the risk behavior response results, the park energy management module uses IoT sensors to monitor and analyze various working environment parameters in real time, adjust air conditioning and lighting, and generate logistics park management records.

[0013] As a further aspect of the present invention, the logistics vehicle dispatch list includes vehicle identification number, park entry time allocation record, and park entry route dispatch parameters; the loading and unloading location allocation result includes loading and unloading point calculation result, loading and unloading time prediction information, and loading and unloading end time record; the park traffic management parameters include traffic flow density data, detour route adjustment result, and congestion warning level; the compliance monitoring record includes employee ID for abnormal attire, information on the type of prohibited equipment, and information on the time of violation; the risk behavior response result includes the risk level of the violation, the record of response measures execution, and the time of warning issuance; and the logistics park management record includes temperature and humidity information, energy consumption data, and air conditioning and lighting equipment adjustment parameters.

[0014] As a further aspect of the present invention, the vehicle entry and exit module includes:

[0015] The vehicle type recording submodule records the license plate number and entry time of each vehicle based on real-time vehicle entry data, identifies the vehicle type, and generates vehicle information identification records.

[0016] The vehicle information analysis submodule analyzes the vehicle's cargo destination and expected dwell time based on the vehicle information identification record, optimizes the vehicle's entry and exit order and route configuration, and generates entry and exit order calculation results.

[0017] Based on the entry and exit order calculation results, the truck scheduling optimization submodule calculates the entry and exit times and routes of multiple vehicles in and out of the logistics park, optimizes waiting time and entry efficiency, and generates a logistics vehicle scheduling list.

[0018] As a further aspect of the present invention, the platform scheduling optimization module includes:

[0019] The loading and unloading location analysis submodule receives the logistics vehicle scheduling list, combines the real-time location of the vehicles and the destination information of the goods, calculates the loading and unloading locations of multiple trucks, and generates loading and unloading location calculation data.

[0020] The flow path analysis submodule calculates the cargo flow path and time based on the loading and unloading location data, adjusts the distribution of loading and unloading locations to optimize the movement distance of goods in the warehouse, and generates loading and unloading location adjustment information;

[0021] The loading and unloading efficiency optimization submodule optimizes cargo turnover time and warehouse storage space utilization based on the loading and unloading location adjustment information, ensuring maximum cargo handling efficiency and generating loading and unloading location allocation results.

[0022] As a further aspect of the present invention, the visitor management module includes:

[0023] The traffic data acquisition submodule collects and analyzes vehicle flow information in the logistics park in real time based on the loading and unloading location allocation results, including vehicle location, speed and flow direction information, and generates a real-time park traffic dataset.

[0024] The park congestion analysis submodule, based on the real-time park traffic dataset, uses the support vector machine algorithm to identify and analyze vehicle congestion, calculate the traffic flow and estimated travel time of congested paths, and generate congestion path identification results.

[0025] Based on the congestion path identification results, the scheduling parameter adjustment submodule adjusts traffic signal parameters and route guidance, optimizes traffic flow and vehicle waiting time, and generates park traffic management parameters.

[0026] As a further aspect of the present invention, the support vector machine algorithm is based on the formula:

[0027]

[0028] Calculate traffic congestion within the logistics park, where f(x) is the output of the decision function, and α... i For the Lagrange multipliers in support vector machines, y i Here, x represents the class label corresponding to the support vector, and x is the input feature vector. i Let t be the i-th feature vector in the training data. i h represents the characteristic value related to vehicle type. i c represents the time period characteristic value of traffic data collection. i e represents the characteristic value of the weather conditions at that time. i w represents the characteristic value influenced by a special event.t For vehicle type weights, w h For time period weights, w c Assuming weather conditions are weighted, w e denoted as event factor weight, b as bias term, N as the total number of support vectors, and i as index label.

[0029] As a further aspect of the present invention, the perimeter security module includes:

[0030] Based on the park traffic management parameters, the employee image acquisition submodule uses image acquisition equipment to monitor the safety equipment worn by staff in key areas in real time and generate safety equipment image data.

[0031] The safety equipment wearing status recognition submodule analyzes the wearing status of safety equipment by park employees based on the safety equipment image data, identifies behaviors of not wearing safety equipment correctly, including safety helmets and goggles, and generates equipment wearing analysis results;

[0032] The wearable information recording submodule records the behavior of employees who did not wear safety equipment correctly, including the time, location and personnel information of the violation, based on the equipment wearing analysis results, and generates a compliance monitoring record.

[0033] As a further aspect of the present invention, the AI ​​recognition and analysis module includes:

[0034] The work behavior monitoring submodule collects real-time image data of workers in key work areas based on the compliance monitoring records, analyzes the behavior and environmental conditions during the work, and generates a real-time work image library.

[0035] The violation identification submodule analyzes the actions of workers in the images based on the real-time operation image library, identifies and records safety violations and dangerous behaviors, and generates behavior analysis results.

[0036] The early warning information sending submodule sends real-time early warning information to managers based on the behavior analysis results, including the location, time and personnel information of the violation, and generates risk behavior response results in response to safety incidents.

[0037] As a further aspect of the present invention, the park energy management module includes:

[0038] Based on the risk behavior response results, the environmental data recording submodule uses IoT sensors to monitor and record environmental information in real time in multiple work and storage areas of the park, including temperature, humidity and light intensity, and generates environmental monitoring data.

[0039] The parameter adjustment calculation submodule calculates the required temperature, humidity and lighting settings for multiple work and storage areas based on the environmental data monitoring records, combined with the type of goods and employee working hours, and generates environmental regulation and control results.

[0040] Based on the environmental regulation control results, the environmental regulation execution submodule takes energy consumption into account, adjusts the parameters of air conditioning and lighting equipment, optimizes the warehousing and working environment of the logistics park, and generates logistics park management records.

[0041] The method for optimizing the safety and efficiency of logistics parks based on IoT and AI, which is executed based on the aforementioned smart park safety and efficiency optimization system based on IoT and AI, includes the following steps:

[0042] S1: Based on real-time vehicle entry data, identify and record vehicle type, cargo destination and expected stay time, calculate the entry and exit sequence and route of multiple vehicles, and generate vehicle entry management parameters.

[0043] S2: Based on the vehicle management parameters entering the park, combined with the real-time location information of the vehicles and the destination data of the goods, calculate the loading and unloading positions of multiple trucks and generate a loading and unloading position adjustment record.

[0044] S3: Based on the loading and unloading position adjustment records, perform real-time analysis of traffic flow data within the park, identify congested areas and adjust the movement paths of multiple trucks to optimize the continuity and efficiency of logistics and generate traffic flow optimization results.

[0045] S4: Based on the traffic flow optimization results, monitor and analyze the safety equipment wearing status of staff in real time, identify safety violations and dangerous behaviors, send early warning information to management personnel, and generate risk monitoring records;

[0046] S5: Based on the aforementioned risk monitoring records, use IoT sensors to monitor and analyze the temperature, humidity, and light intensity within the park in real time, consider energy consumption, adjust the parameters of air conditioning and lighting equipment, and generate logistics park management records.

[0047] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0048] In this invention, by analyzing information on vehicles and goods entering the park, the order of vehicle entry and exit and route configuration are adjusted to improve entry efficiency and reduce waiting time. Real-time vehicle location information is used to calculate the loading and unloading location of goods, reducing the dwell time of goods during transfer. By analyzing real-time traffic flow data, the optimal logistics path is calculated to optimize traffic flow efficiency within the park. Real-time monitoring of staff behavior and rapid identification of dangerous behaviors and personnel wearing equipment incorrectly improves accident response speed and safety management level. By intelligently adjusting facilities such as air conditioning and lighting, energy utilization and work efficiency are improved. Attached Figure Description

[0049] Figure 1 This is a system flowchart of the present invention;

[0050] Figure 2 This is a schematic diagram of the system framework of the present invention;

[0051] Figure 3 This is a flowchart of the vehicle entry and exit module of the present invention;

[0052] Figure 4 This is a flowchart of the platform scheduling optimization module of the present invention;

[0053] Figure 5 This is a flowchart of the visitor management module of the present invention;

[0054] Figure 6 This is a flowchart of the perimeter security module of the present invention;

[0055] Figure 7 This is a flowchart of the AI ​​recognition and analysis module of the present invention;

[0056] Figure 8 This is a flowchart of the park energy consumption management module of the present invention;

[0057] Figure 9 This is a schematic diagram of the method steps of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0059] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0060] Example

[0061] Please see Figures 1 to 2 This invention provides a technical solution: a smart park security and efficiency optimization system based on the Internet of Things and AI, comprising:

[0062] The vehicle entry and exit module analyzes and identifies vehicle type, cargo destination and expected stay time based on real-time vehicle entry data, calculates the entry and exit sequence and route configuration of multiple vehicles, and obtains a logistics vehicle scheduling list.

[0063] The platform scheduling optimization module calculates the loading and unloading positions of vehicles based on the logistics vehicle scheduling list, combined with the real-time location of vehicles and the destination of goods, and generates loading and unloading position allocation results.

[0064] Based on the loading and unloading location allocation results, the visitor management module collects real-time traffic flow data within the logistics park, calculates the optimal logistics route, and generates park traffic management parameters.

[0065] The perimeter security module uses image acquisition equipment to collect and analyze the safety equipment wearing status of staff in real time based on the park's traffic management parameters, and generates compliance monitoring records.

[0066] The AI ​​recognition and analysis module, based on compliance monitoring records, collects images of workers' operations in real time, identifies safety violations and dangerous behaviors, sends early warning information to managers, and generates risk behavior response results.

[0067] Based on the results of risk behavior response, the park's energy management module uses IoT sensors to monitor and analyze various working environment parameters in real time, adjust air conditioning and lighting, and generate logistics park management records.

[0068] The logistics vehicle dispatch list includes vehicle identification number, entry time allocation record, and entry route dispatch parameters. The loading and unloading location allocation results include loading and unloading point calculation results, loading and unloading time prediction information, and loading and unloading end time records. The park traffic management parameters include traffic flow density data, detour route adjustment results, and congestion warning level. The compliance monitoring record includes employee IDs for abnormal attire, information on the type of prohibited equipment, and information on the time of the violation. The risk behavior response results include the risk level of the violation, the record of the response measures implemented, and the time of the warning. The logistics park management record includes temperature and humidity information, energy consumption data, and air conditioning and lighting equipment adjustment parameters.

[0069] Please see Figure 2 and Figure 3 The vehicle entry and exit module for the park includes:

[0070] The vehicle type recording submodule records the license plate number and entry time of each vehicle based on real-time vehicle entry data, identifies the vehicle type, and generates vehicle information identification records. The specific process is as follows:

[0071] In the vehicle type recording submodule, based on real-time vehicle entry data, information for each vehicle entering the logistics park is recorded, including license plate number and entry time. Using a vehicle recognition system, the types of multiple vehicles are identified based on license plate and vehicle model characteristics. The impact of vehicle type on park traffic flow is analyzed, employing a vehicle type recognition model with the following formula: Among them, T i x represents the vehicle type index of the i-th vehicle. ik w represents the value of vehicle i in the k-th feature. k This represents the weighting coefficient of the corresponding vehicle model features, and generates vehicle information recognition records.

[0072] The vehicle information analysis submodule analyzes the vehicle's cargo destination and expected dwell time based on vehicle information identification records, optimizes the vehicle's entry and exit sequence and route configuration, and generates the entry and exit sequence calculation results. The specific process is as follows:

[0073] In the vehicle information analysis submodule, based on vehicle information identification records, vehicles are classified and managed by analyzing the cargo destination and expected dwell time of each vehicle. Data for each vehicle type is summarized, and the average dwell time and optimal entry / exit route for each type are calculated using a vehicle flow analysis model, with the formula: S i =α×D i +β×T i , among which, S i Let D be the optimal order value for vehicle i. i Indicates destination priority, T iThe expected dwell time is represented by α and β, which are the weighting parameters of destination and time, respectively. The optimal vehicle entry and exit order is calculated, and the entry and exit order calculation results are generated.

[0074] The truck scheduling optimization submodule calculates the entry and exit order results, calculates the entry and exit times and routes of multiple vehicles in and out of the logistics park, optimizes waiting time and entry efficiency, and generates a logistics vehicle scheduling list. The specific process is as follows:

[0075] In the truck scheduling optimization submodule, based on the entry and exit order calculation results, the time and route for multiple vehicles to enter and exit the logistics park are calculated. A dynamic scheduling optimization model is used, with the following formula: Among them, O i p represents the overall optimization metric for vehicle i. j q represents the influence of the j-th time or route parameter. ij This indicates the performance of vehicle i under this parameter. Based on the entry efficiency and waiting time of various vehicles, the vehicle scheduling parameters are optimized to generate a logistics vehicle scheduling list.

[0076] Please see Figure 2 and Figure 4 The platform scheduling optimization module includes:

[0077] The loading and unloading location analysis submodule receives the logistics vehicle dispatch list, combines the real-time location of the vehicles and the destination information of the goods, calculates the loading and unloading locations of multiple trucks, and generates loading and unloading location calculation data. The specific process is as follows:

[0078] In the loading and unloading location analysis submodule, a logistics vehicle scheduling list is received. The real-time location of each vehicle is obtained using a vehicle tracking system. Combined with the vehicle's cargo destination information, the optimal loading and unloading location for each truck is calculated using a geographic information system. This includes distance and time cost analysis for each truck's designated loading and unloading locations, employing a location optimization model with the following formula: Among them, L i c represents the cost of the optimal loading / unloading location for truck i. k d represents the cost coefficient for the k-th location. ik Generate loading and unloading location calculation data for the distance of truck i to the k-th location.

[0079] The flow path analysis submodule calculates data based on loading and unloading locations, analyzes the flow path and time of goods, adjusts the distribution of loading and unloading locations to optimize the movement distance of goods in the warehouse, and generates loading and unloading location adjustment information. The specific process is as follows:

[0080] In the flow path analysis submodule, based on loading and unloading location calculation data, the flow path and required time of goods are analyzed. Through the goods path planning system, the loading and unloading locations of multiple trucks are optimized and adjusted to reduce the movement distance within the warehouse. The formula is: Among them, P i For the path efficiency of truck i, e j f represents the efficiency parameter of the j-th path node. ij γ represents the frequency at which truck i passes through the j-th node, and γ is the adjustment coefficient. By optimizing the selection of path nodes and frequency allocation, the loading and unloading positions are adjusted, generating loading and unloading position adjustment information.

[0081] The loading and unloading efficiency optimization submodule optimizes cargo turnover time and warehouse storage space utilization based on loading and unloading location adjustment information to ensure maximum cargo handling efficiency. The specific process for generating loading and unloading location allocation results is as follows:

[0082] In the loading and unloading efficiency optimization submodule, based on the loading and unloading location adjustment information, the utilization rate of cargo turnover time and warehouse storage space is optimized by applying the cargo turnover efficiency model, the formula of which is: Among them, E i For the loading and unloading efficiency of truck i, g k h represents the efficiency parameter of the k-th storage region. ik δ represents the amount of cargo stored in truck i in region k, and δ is the efficiency adjustment factor. By adjusting the loading and unloading positions of trucks and the distribution of cargo storage, the processing efficiency is maximized, and the loading and unloading position allocation results are generated.

[0083] Please see Figure 2 and Figure 5 The visitor management module includes:

[0084] The traffic data collection submodule collects and analyzes vehicle flow information in the logistics park in real time based on the loading and unloading location allocation results, including vehicle location, speed and flow direction information, and generates a real-time park traffic dataset. The specific process is as follows:

[0085] In the traffic data acquisition submodule, a real-time data acquisition framework is built based on the loading and unloading location allocation results. Sensors and GPS tracking devices are used to monitor the location, speed, and flow direction of vehicles within the logistics park in real time. This includes collecting and storing the speed and location data of each vehicle in a central database. The data analysis module processes the target information, calculates traffic density and movement trends, and utilizes a traffic flow data analysis model, with the following formula: Among them, V i The comprehensive traffic data index representing the i-th vehicle, v ij s represents the measured value of vehicle i at the j-th parameter, including speed and direction. jThe standardized coefficients of the target parameters are used to analyze traffic flow dynamics and generate a real-time park traffic dataset.

[0086] The park congestion analysis submodule is based on real-time park traffic datasets and uses the support vector machine algorithm to identify and analyze vehicle congestion, calculate traffic flow and estimated travel time for congested paths, and generate congestion path identification results. The specific process is as follows:

[0087] In the park congestion analysis submodule, based on the real-time park traffic dataset, vehicle flow information within the park is collected, including vehicle speed, location, and entrance / exit usage. The collected data is preprocessed to remove erroneous and incomplete data records, and vehicle timestamps and location information are standardized. Through a defined data interface, traffic flow changes at multiple traffic nodes are monitored in real time, and traffic flow and average speed within a target time window are calculated. Statistical data on traffic flow and speed across multiple road segments are analyzed. By comparing the deviation with historical data from the same period, congestion points are identified, and based on the current traffic flow status, the estimated travel time of congested paths is predicted, generating congestion path identification results.

[0088] The Support Vector Machine algorithm, according to the formula:

[0089]

[0090] Calculate traffic congestion within the logistics park, where f(x) is the output of the decision function used to predict vehicle congestion, and α... i The Lagrange multipliers in a support vector machine are used to adjust the contribution of each support vector, y. i Here, x represents the class label corresponding to the support vector, and x is the input feature vector, including relevant features from real-time traffic data. i Let t be the i-th feature vector in the training data. i h represents the characteristic value related to vehicle type. i c represents the time period characteristic value of traffic data collection. i e represents the characteristic value of the weather conditions at that time. i w represents the characteristic value influenced by a special event. t Weighting by vehicle type adjusts the impact of different vehicle types on congestion. h The time period weights are used to adjust the impact of data from different time periods on congestion prediction. c Weighting weather conditions to adjust for the impact of weather changes on traffic flow, w e is the weight of the event factors, which adjusts the impact of special events on traffic flow; b is the bias term, used to adjust the decision boundary in the decision function; N is the total number of support vectors; and i is the index label.

[0091] The specific execution process of the formula is as follows: Traffic data is acquired through a real-time monitoring system, including vehicle type, time period, weather conditions, and information on any special events. Support vector machines are then used to integrate the vehicle type data. i Time period h i Weather conditions c i The impact of the event and its corresponding weight w t w h w c w e By substituting parameters into the formula, congestion can be predicted, which can then guide vehicle scheduling and traffic management strategies within logistics parks, optimize traffic flow, and reduce congestion.

[0092] The scheduling parameter adjustment submodule adjusts traffic signal parameters and route guidance based on congestion path identification results, optimizes traffic flow and vehicle waiting time, and generates park traffic management parameters. The specific process is as follows:

[0093] In the scheduling parameter adjustment submodule, based on the congestion path identification results, traffic signal parameters and route guidance are adjusted to optimize traffic flow and vehicle waiting time. This includes analyzing congestion results and traffic flow data, adjusting the timing of traffic lights and route guidance signs, and employing a traffic flow optimization model with the following formula: Among them, J i y represents the traffic management optimization index at the i-th adjustment point. j λ is the adjustment parameter for the j-th signal or guide. ij The impact coefficient on traffic flow at point i is adjusted to target the overall adjustment effect, and α is the weighting factor for the overall adjustment effect. This optimizes traffic congestion and waiting time, and generates traffic management parameters for the park.

[0094] Please see Figure 2 and Figure 6 The perimeter security module includes:

[0095] The employee image acquisition submodule uses image acquisition equipment to monitor the safety equipment worn by staff in key areas in real time based on park traffic management parameters and generates safety equipment image data. The specific process is as follows:

[0096] In the employee image acquisition submodule, based on park traffic management parameters, image acquisition equipment is used to monitor the wearing of safety equipment by staff in key areas of the park in real time. The equipment is deployed at entrances, work areas, and hazardous areas, automatically capturing improper attire, including using image recognition technology to automatically detect the wearing of safety helmets and goggles, and the corresponding equipment color and type. The image data analysis model is used, with the following formula: Among them, I i a represents the image recognition index of employee i's safety equipment. kThe r represents the identification parameter of security equipment k. ik Based on the recognition result of employee i in image k, generate safety equipment image data.

[0097] The equipment wearing identification submodule analyzes the wearing status of safety equipment by the image data of safety equipment, identifies the behavior of not wearing safety equipment correctly, including safety helmets and goggles, and generates equipment wearing analysis results. The specific process is as follows:

[0098] In the safety equipment wear identification submodule, based on safety equipment image data, image processing algorithms are used to analyze the wearing status of safety equipment among park employees, identify and mark employees who are not wearing safety equipment correctly, and use a wear analysis model, the formula of which is: Among them, W i The degree of equipment wearing violation by employee i, b j For the violation weight of security equipment type j, s ij Let γ be the identification score of employee i for equipment j, and γ be the adjustment coefficient. This identifies the behavior of not wearing equipment as required and generates equipment wearing analysis results.

[0099] The wearable information recording submodule records employee behavior of not wearing safety equipment correctly based on the equipment wearing analysis results, including the time, location and personnel information of the violation, and the specific process for generating compliance monitoring records is as follows;

[0100] In the wearability information recording submodule, based on the equipment wearing analysis results, the behavior of employees who did not wear safety equipment correctly is recorded, along with the time, location, and personnel information of the violation. The identified violations are matched against the employee database to determine the violator's identity. A compliance monitoring model is used, with the following formula: Among them, C i For employee i, the compliance monitoring index, d l Indicates the importance of the monitoring parameter l, e il Generate compliance monitoring records for employee i's violations under parameter l.

[0101] Please see Figure 2 and Figure 7 The AI ​​recognition and analysis module includes:

[0102] The work behavior monitoring submodule collects real-time image data of workers in key work areas based on compliance monitoring records, analyzes their behavior and environmental conditions during the work period, and generates a real-time work image library. The specific process is as follows:

[0103] In the work behavior monitoring submodule, based on compliance monitoring records, multiple high-resolution cameras are used to collect real-time image data of key work areas, capturing worker postures, tool usage, and interactions with the surrounding environment. The image data undergoes preprocessing, including image denoising, contrast enhancement, and edge detection, to improve the accuracy of subsequent analysis. The image processing formula is used: I processed =α·(I original *G σ )+β, where I processed For the processed image, I original For the original image data, G σ A Gaussian filter is used for image smoothing, α is the contrast adjustment coefficient, and β is the brightness adjustment parameter, generating a real-time job image library.

[0104] The violation identification submodule analyzes the actions of workers in real-time operation images based on the image library, identifies and records safety violations and dangerous behaviors, and generates behavior analysis results. The specific process is as follows:

[0105] In the violation recognition submodule, based on a real-time work image library, different work actions are distinguished by the movements of workers in the images, and safety violations and dangerous behaviors are identified. A behavior recognition formula is applied to each image frame. Where R represents the risk score of the violation, N is the number of action categories, and w i Let f be the risk weight for the i-th behavioral category. i (X) is the recognition function for the corresponding behavior, where X is the input image data, and the behavior analysis results are generated.

[0106] The early warning information sending submodule sends real-time early warning information to managers based on the behavior analysis results, including the location, time and personnel information of the violation, and responds to safety incidents, generating risk behavior response results. The specific process is as follows:

[0107] In the early warning information sending submodule, based on the behavior analysis results, the information of the violation, including location, time and personnel information, is sent to the monitoring equipment of the management personnel through the integrated communication interface. The sending process adopts the information encoding formula: M = encode(T,L,P), where M is the message to be sent, encode is the encoding function, T is the violation time, L is the location information, and P is the personnel identification information, which helps the management personnel respond to safety incidents and generate risk behavior response results.

[0108] Please see Figure 2 and Figure 8 The park's energy management module includes:

[0109] The environmental data recording submodule, based on the risk behavior response results, uses IoT sensors to monitor and record environmental information in real time in multiple work and storage areas of the park, including temperature, humidity and light intensity. The specific process for generating environmental monitoring data is as follows:

[0110] In the environmental data recording submodule, based on the risk behavior response results, multiple IoT sensors are deployed to monitor the environmental information of the work and storage areas within the park in real time, collecting data on temperature, humidity, and light intensity, and using the environmental monitoring formula: E recorded =s T ·T+s H ·H+s L ·L, where E recorded For the environmental data to be recorded, s T Here, T represents the calibration coefficient of the temperature sensor, used to convert the sensor reading to the actual temperature value; s represents the actual temperature reading, used to measure the temperature conditions in the environment. H Here, is the calibration coefficient for the humidity sensor, used to convert sensor readings to actual humidity values; H is the actual humidity reading, used to measure humidity conditions in the environment; and s... L The calibration coefficient for the light intensity sensor is used to convert the sensor reading to the actual light intensity. L is the actual light intensity reading, used to measure the lighting conditions in the environment and generate environmental monitoring data.

[0111] The parameter adjustment calculation submodule calculates the required temperature, humidity, and lighting settings for multiple work and storage areas based on environmental data monitoring records, combined with cargo type and employee working hours, and generates environmental regulation and control results. The specific process is as follows:

[0112] In the parameter adjustment calculation submodule, based on environmental data monitoring records and considering the type of stored goods and employee working hours, the ideal temperature, humidity, and lighting intensity for the storage and work areas are calculated using the environmental conditioning formula: C adjust =γ·(a·T) ideal +b·H ideal +c·L ideal ), where C adjust The adjusted environmental control parameters are: γ, the environmental adjustment coefficient used to adjust the output of temperature, humidity, and lighting according to actual needs; a, the temperature adjustment weight used to adjust the temperature setting according to the type of goods and working environment requirements; and T. ideal 'b' represents the ideal temperature setting, indicating the optimal temperature required for the target goods and working hours. 'b' represents the humidity adjustment weight, used to adjust the humidity setting based on goods storage requirements and employee comfort. ideal The ideal humidity setting represents the optimal humidity required for the target goods and working environment. c represents the lighting adjustment weight, used to adjust the lighting intensity according to the specific needs of the work area. Lideal Ideal lighting intensity is used to represent the required light level for the work area under target conditions, generating environmental conditioning and control results.

[0113] The environmental regulation execution submodule, based on the environmental regulation control results and considering energy consumption, adjusts the parameters of air conditioning and lighting equipment, optimizes the warehousing and working environment of the logistics park, and generates logistics park management records. The specific process is as follows:

[0114] In the environmental regulation execution submodule, based on the environmental regulation control results and considering energy consumption, the operating parameters of air conditioning and lighting equipment are adjusted through an intelligent regulation system. This includes setting air conditioning temperature and humidity to optimal values, adjusting lighting intensity to meet current light requirements, and minimizing energy consumption using energy optimization formulas. optimized =δ·(d·E ac +e·E light ), where O optimized The optimized energy consumption value is given by δ, which is the energy efficiency coefficient used to adjust the equipment operation and optimize energy consumption, and d, which is the air conditioning energy consumption factor used to calculate the energy consumption of the air conditioning equipment after adjustment. ac E represents the initial energy consumption of the air conditioning equipment, used to measure the energy usage before adjustment; e represents the energy consumption factor of the lighting equipment, used to calculate the energy consumption of the lighting equipment after adjustment. light It measures the initial energy consumption of lighting equipment, measures energy usage before adjustment, optimizes the warehousing and working environment of the logistics park, and generates logistics park management records.

[0115] Please see Figure 9 The method for optimizing the safety and efficiency of logistics parks based on IoT and AI is implemented based on the aforementioned IoT and AI-based logistics park safety and efficiency optimization system, and includes the following steps:

[0116] S1: Based on real-time vehicle entry data, identify and record vehicle type, cargo destination and expected stay time, calculate the entry and exit sequence and route of multiple vehicles, and generate vehicle entry management parameters.

[0117] S2: Based on the vehicle management parameters entering the park, combined with the real-time location information of the vehicles and the destination data of the goods, calculate the loading and unloading positions of multiple trucks and generate loading and unloading position adjustment records.

[0118] S3: Based on loading and unloading location adjustment records, perform real-time analysis of traffic flow data within the park, identify congested areas and adjust the movement paths of multiple trucks to optimize the continuity and efficiency of logistics and generate traffic flow optimization results.

[0119] S4: Based on traffic flow optimization results, monitor and analyze the wearing status of safety equipment by staff in real time, identify safety violations and dangerous behaviors, send early warning information to management personnel, and generate risk monitoring records;

[0120] S5: Based on risk monitoring records, use IoT sensors to monitor and analyze the temperature, humidity and light intensity in the park in real time, take energy consumption into account, adjust the parameters of air conditioning and lighting equipment, and generate logistics park management records.

[0121] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A smart park security and efficiency optimization system based on Internet of Things and AI, characterized in that, The system comprises: The vehicle access module analyzes and identifies the vehicle type, cargo destination and expected stay time based on real-time vehicle access data, calculates the access sequence and route configuration of multiple vehicles, and obtains a logistics vehicle scheduling list; The platform scheduling optimization module calculates the loading and unloading location of the vehicle based on the logistics vehicle scheduling list, combines the real-time location of the vehicle and the cargo destination, and generates a loading and unloading location allocation result; The perimeter security module uses image acquisition equipment to collect and analyze the safety equipment wearing condition of the staff in real time based on the traffic management parameters of the park, and generates compliance monitoring records; The AI recognition analysis module identifies the safety violations and dangerous behaviors of the staff based on the compliance monitoring records, collects the work images of the staff in real time, sends early warning information to the management personnel, and generates a risk behavior response result; The park energy consumption management module uses Internet of Things sensors to monitor and analyze various working environment parameters in real time based on the risk behavior response result, and records the park energy consumption usage. 2.The IoT and Al-based smart park security and efficiency optimization system according to claim 1, wherein, The logistics vehicle scheduling list includes vehicle identification number, access time allocation record, and access route scheduling parameter, the loading and unloading location allocation result includes loading and unloading point calculation result, loading and unloading time prediction information, and loading and unloading end time record, the park traffic management parameter includes traffic density data, detour route adjustment result, and congestion warning level, the compliance monitoring record includes abnormal wearing employee number, rule violation equipment type information, and rule violation time point information, the risk behavior response result includes rule violation risk level, response measure execution record, and early warning time point, and the logistics park management record includes temperature and humidity information, energy consumption data, and air conditioning and lighting device adjustment parameter. 3.The IoT and Al-based smart park security and efficiency optimization system according to claim 1, wherein, The vehicle access module comprises: The vehicle type record sub-module records the license plate number and access time of each vehicle based on real-time vehicle access data, identifies the vehicle type, and generates a vehicle information identification record; The vehicle information analysis sub-module analyzes the cargo destination and expected stay time of the vehicle based on the vehicle information identification record, optimizes the access sequence and route configuration of the vehicle, and generates an access sequence calculation result; The truck scheduling optimization sub-module calculates the access time and route of multiple vehicles into and out of the logistics park based on the access sequence calculation result, optimizes the waiting time and access efficiency, and generates a logistics vehicle scheduling list.

4. The IoT and Al based smart campus security and efficiency optimization system according to claim 1, wherein, The platform scheduling optimization module comprises: The loading and unloading site analysis sub-module receives the logistics vehicle scheduling list, combines the real-time location of the vehicle and the cargo destination information, calculates the loading and unloading location of multiple trucks, and generates loading and unloading location calculation data; The flow path analysis sub-module analyzes the cargo flow path and time based on the loading and unloading location calculation data, adjusts the loading and unloading location distribution to optimize the movement distance of the cargo in the warehouse, and generates loading and unloading location adjustment information; The loading and unloading efficiency optimization sub-module optimizes the cargo flow time and warehouse storage space utilization based on the loading and unloading location adjustment information, ensures the maximum cargo handling efficiency, and generates a loading and unloading location allocation result. 5.The IoT and Al-based smart park security and efficiency optimization system according to claim 1, wherein, The visitor management module comprises: The traffic data collection submodule collects and analyzes the traffic information in the logistics park in real time according to the loading and unloading location allocation result, including the position, speed and flow direction information of the vehicle, and generates a real-time park traffic data set; The park congestion analysis submodule identifies and analyzes the vehicle congestion situation based on the real-time park traffic data set, calculates the traffic flow and expected travel time of the congestion path, and generates a congestion path identification result by using a support vector machine algorithm; The scheduling parameter adjustment submodule adjusts the traffic signal parameters and route guidance based on the congestion path identification result, optimizes the traffic flow and vehicle waiting time, and generates park traffic management parameters. 6.The IoT and Al-based smart park security and efficiency optimization system according to claim 5, wherein, The support vector machine algorithm is according to the formula: Traffic congestion in the logistics park is calculated, where f(x) is the output of the decision function, a i is the Lagrange multiplier in the support vector machine, y i is the class label corresponding to the support vector, x is the input feature vector, x i is the i-th feature vector in the training data, t i is the feature value related to the vehicle type, h i is the time period feature value of traffic data collection, c i is the weather condition feature value at that time, e i is the feature value of special event influence, w t is the vehicle type weight, w h is the time period weight, w c is the weather condition weight, w e is the event factor weight, b is the bias term, and N is the total number of support vectors, i is the index label.

7. The IoT and Al-based smart park security and efficiency optimization system according to claim 1, wherein, The perimeter security module includes: The employee image collection submodule uses image collection equipment to monitor the safety equipment wearing situation of the staff in the key area in real time based on the park traffic management parameters, and generates safety equipment image data; The wearing situation identification submodule analyzes the wearing situation of the safety equipment of the park employees based on the safety equipment image data, identifies the behavior of not correctly wearing safety equipment, including safety helmets and goggles, and generates an equipment wearing analysis result; The wearing information recording submodule records the behavior of the staff who do not correctly wear safety equipment according to the equipment wearing analysis result, including the information of violation time, place and personnel, and generates a compliance monitoring record. 8.The IoT and Al-based smart park security and efficiency optimization system according to claim 1, wherein, The AI identification analysis module includes: The work behavior monitoring submodule collects image data of the staff in the key work area in real time according to the compliance monitoring record, analyzes the behavior and environmental situation during work, and generates a real-time work image library; The violation behavior identification submodule analyzes the actions of the staff in the image based on the real-time work image library, identifies and records safety violation behaviors and dangerous behaviors, and generates a behavior analysis result; The early warning information sending submodule sends real-time early warning information to the management personnel according to the behavior analysis result, including the location, time and personnel information of the violation behavior, and responds to safety accidents, and generates a risk behavior response result. 9.The IoT and Al-based smart park security and efficiency optimization system according to claim 1, wherein, The park energy consumption management module includes: The environmental data recording submodule uses Internet of Things sensors to monitor and record the environmental information of multiple work and storage areas in the park in real time based on the risk behavior response result, including temperature, humidity and illumination intensity, and generates environmental monitoring data; The adjustment parameter calculation submodule calculates the temperature, humidity and lighting settings required by multiple work and storage areas based on the environmental data monitoring record, combined with the type of goods and the working time of the staff, and generates an environmental regulation control result; The environmental regulation execution submodule adjusts the parameters of air conditioning and lighting equipment according to the environmental regulation control result, considering energy consumption, optimizes the storage and working environment of the logistics park, and generates a logistics park management record.

10. A method for safety and efficiency optimization of a smart park based on Internet of Things and AI, characterized in that, The Internet of Things and AI-based intelligent park safety and efficiency optimization system according to any one of claims 1-9, comprising the following steps: Based on the real-time vehicle entry data, the type, destination and expected stay time of the vehicle are identified and recorded, the entry and exit order and route of multiple vehicles are calculated, and entry vehicle management parameters are generated; Based on the vehicle management parameters, combined with the real-time position information of the vehicle and the cargo destination data, the loading and unloading positions of multiple trucks are calculated, and a loading and unloading position adjustment record is generated; Based on the loading and unloading position adjustment record, real-time analysis is performed on the traffic flow data in the park, congestion areas are identified, and the moving path of multiple trucks is adjusted to optimize the continuity and efficiency of logistics, and a traffic flow optimization result is generated; Based on the traffic flow optimization result, the safety equipment wearing condition of the staff is monitored and analyzed in real time, safety violations and dangerous behaviors are identified, and warning information is sent to the management personnel, and a risk monitoring record is generated; Based on the risk monitoring record, the temperature, humidity and light intensity in the park are monitored and analyzed in real time by using Internet of Things sensors, considering energy consumption, adjusting the parameters of air conditioning and lighting equipment, and generating a logistics park management record.

Images