Smart city Internet of Things equipment data acquisition management platform and implementation method thereof

Abstract

The invention relates to the technical field of Internet of Things data management, in particular to a smart city Internet of Things equipment data acquisition management platform and an implementation method thereof. The method comprises the following steps: deploying Internet of Things sensing equipment in a garbage can, and constructing an Internet of Things front-end sensing network; the method comprises the following steps: acquiring data of an incremental throwing event in a garbage can through an Internet of Things front-end sensing network, and calculating a state emergency degree in the garbage can to obtain a dynamic urgency degree index; performing supervision weighted clearance urgency degree processing according to the dynamic urgency degree index, and dynamically planning a theoretical optimal receiving and transporting sequence of the target receiving and transporting vehicle; and carrying out segmented navigation instruction processing on the theoretical optimal collection and transportation sequence so as to manage the collection and transportation operation of the intelligent garbage can. The garbage state is monitored in real time by deploying a multi-sensor network in the garbage can, garbage thermal decomposition dynamic monitoring and regional differentiation clearance priority intelligent sorting are achieved, and the whole process of environmental sanitation operation is accurately guided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things data management, and particularly relates to a smart city Internet of Things device data collection management platform and an implementation method thereof. BACKGROUND

[0002] The main problems of urban garbage collection and transportation management are as follows: garbage can overflow frequently, causing environmental health hazards; the collection and transportation routes are unreasonable, causing resource waste; the collection and transportation frequency does not match the actual demand, causing low efficiency, etc. Especially in densely populated areas and commercial prosperous areas, the garbage generation fluctuates greatly, and the garbage can easily overflows during the peak period, which not only affects the city appearance, but also causes public health problems and citizen complaints. According to statistics, the number of complaints of environmental health problems caused by garbage can overflow in large and medium-sized cities reaches tens of thousands per year, and the cost of related environmental sanitation emergency treatment accounts for 15%-20% of the total cost of garbage collection and transportation. The existing intelligent garbage monitoring system usually measures the filling height of the garbage can by using a single parameter, and infers the filling state of the garbage can based on the filling height. However, this simple technical solution has obvious defects: only measuring the surface height cannot accurately reflect the real capacity utilization rate of the garbage can. Because the stacking state, density and distribution of garbage are extremely uneven, the situation of "false high" or "false low" may occur. For example, bulky and light garbage (such as packaging foam and paper boxes) will quickly occupy the upper space of the garbage can, triggering a high filling rate alarm; on the contrary, small and high-density garbage (such as kitchen waste and construction waste) will be concentrated at the bottom, and the filling height is not high but the actual capacity is close to the limit. SUMMARY

[0003] Therefore, the present application provides a smart city Internet of Things device data collection management platform and an implementation method thereof to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned purpose, an implementation method of a smart city Internet of Things device data collection management platform comprises the following steps:

[0005] Step S1: deploying an Internet of Things sensing device in the garbage can to build an Internet of Things front-end sensing network; collecting the data of the incremental throwing event in the garbage can through the Internet of Things front-end sensing network, and analyzing the structural voxel changes caused by the incremental throwing event to generate garbage three-dimensional structural voxel data;

[0006] Step S2: calculating the total filling rate of garbage according to the garbage three-dimensional structural voxel data; evaluating the dynamic thermal decomposition degree data according to the garbage three-dimensional structural voxel data, and calculating the in-can state urgency according to the total filling rate of garbage to obtain a dynamic urgency index; performing supervision weighted collection and transportation urgency processing according to the dynamic urgency index, and then identifying a set of to-be-processed stations;

[0007] Step S3: extracting the to-be-visited site coordinate data in the to-be-processed site cluster, and dynamically planning a theoretical optimal collection and transportation sequence of the target collection and transportation vehicle;

[0008] Step S4: processing the theoretical optimal collection and transportation sequence by segmented navigation instruction, and adding a single-point fine operation note through the garbage three-dimensional structure voxel data, so as to manage the intelligent garbage can collection and transportation operation.

[0009] Preferably, the present application also provides a smart city Internet of Things equipment data collection management platform, comprising a data layer, a service layer and an application layer, the output end of the data layer is provided with the service layer, the output end of the service layer is provided with the application layer, the data layer is used for collecting smart city information and performing analysis, the service layer is used for providing different services to meet the access demand of the application layer, the application layer realizes the access of the intelligent terminal equipment through an access interface, and the data layer is used for executing the implementation method of the smart city Internet of Things equipment data collection management platform as described above.

[0010] The Internet of Things data collection module is used for deploying Internet of Things sensing equipment in the garbage can, constructing an Internet of Things front-end sensing network, collecting the data of the incremental throwing events in the garbage can through the Internet of Things front-end sensing network, analyzing the structure voxel changes caused by the incremental throwing events in the garbage can, and generating the garbage three-dimensional structure voxel data.

[0011] The dynamic risk assessment module is used for calculating the total filling rate of garbage according to the garbage three-dimensional structure voxel data, evaluating the dynamic thermal decomposition degree data according to the garbage three-dimensional structure voxel data, calculating the in-can state emergency degree according to the total filling rate of garbage, obtaining a dynamic urgency index, performing supervision weighted clean-up urgency processing according to the dynamic urgency index, and then identifying the to-be-processed site cluster.

[0012] The collection and transportation path planning module is used for extracting the to-be-visited site coordinate data in the to-be-processed site cluster, and dynamically planning a theoretical optimal collection and transportation sequence of the target collection and transportation vehicle.

[0013] The intelligent operation management module is used for processing the theoretical optimal collection and transportation sequence by segmented navigation instruction, and adding a single-point fine operation note through the garbage three-dimensional structure voxel data, so as to realize the intelligent management of the garbage can collection and transportation in the smart city.

[0014] The present application realizes accurate evaluation of the real filling state of the garbage can by precisely capturing the three-dimensional structure data of garbage instead of simple surface height measurement. This stereoscopic monitoring mode based on the front-end sensing network of the Internet of Things can effectively distinguish between light bulky and high-density small-volume garbage, accurately reflect the capacity utilization, and make the early warning mechanism more reliable. Through comprehensive analysis of dynamic thermal decomposition degree data and filling rate, the actual emergency degree of the internal state of the garbage can can be intelligently judged, avoiding misjudgment caused by single parameter judgment, and greatly reducing the sanitation emergency treatment cost by about 15%-20%. The supervision weighted collection and transportation processing based on the dynamic urgency index of the platform realizes the change from "fixed cycle collection and transportation" to "on-demand collection and transportation", greatly improving the collection and transportation efficiency. Through intelligent identification and clustering analysis of the to-be-processed station, combined with the theoretically optimal collection and transportation sequence of dynamic programming, the garbage collection and transportation route is significantly optimized, reducing the empty mileage and fuel consumption, reducing carbon emissions and operating costs. The fine segmented navigation instruction and operation note function provide clear operation guidance for frontline sanitation workers, improving operation quality and efficiency. In addition, the platform provides a comprehensive view of garbage generation, distribution pattern and treatment efficiency for city managers through precise data collection and analysis, supporting scientific decision-making and resource optimization. The intelligent prediction function of the platform can predict the garbage generation during peak period according to historical data patterns, adjust the collection and transportation plan in advance, and effectively prevent overflow events. In densely populated areas and business prosperous areas, by accurately responding to the large fluctuation of garbage generation, the urban environmental sanitation and city appearance are ensured, the public complaints are reduced, and the public satisfaction is improved. Therefore, the implementation method of the smart city Internet of Things equipment data acquisition and management platform of the present application constructs garbage three-dimensional structure voxel data through a multi-sensor front-end sensing network, analyzes garbage decomposition activity combined with infrared thermal imaging, evaluates risk urgency using meteorological data and environmental parameters, realizes regional differentiated collection and transportation priority ranking, and estimates dynamic unloading impact value based on internal structure cavity identification and centroid offset analysis, generates fine operation guidance, thereby realizing intelligent management of the whole process of garbage monitoring, evaluation, scheduling, collection and transportation. BRIEF DESCRIPTION OF DRAWINGS

[0015] Fig. 1 The step flowchart of the implementation method of the smart city Internet of Things equipment data acquisition and management platform of the present application is shown in the figure.

[0016] Fig. 2 The voxel activity score fold line diagram in the smart city Internet of Things equipment data acquisition and management platform is shown in the figure.

[0017] The implementation of the purpose of the present application, functional characteristics and advantages will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0018] The technical method of the present application will be described clearly and completely in combination with the drawings. Obviously, the described embodiments are part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0019] In addition, the drawings are only schematic illustrations of the present application, and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated description thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities, which do not necessarily have to correspond to physically or logically independent entities. The functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0020] It should be understood that although the terms "first", "second" and the like can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of the exemplary embodiments, a first element can be called a second element, and similarly a second element can be called a first element. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0021] To achieve the above-mentioned purpose, please refer to Figs. 1-2 The present application provides an implementation method of a smart city Internet of Things device data acquisition and management platform, comprising the following steps:

[0022] Step S1: deploying Internet of Things sensing devices inside the garbage can to build an Internet of Things front-end sensing network; collecting data of incremental throwing events in the garbage can through the Internet of Things front-end sensing network, and analyzing the changes of structural voxels in the garbage can caused by the incremental throwing events to generate garbage three-dimensional structural voxel data;

[0023] In the embodiment of the present application, a plurality of Internet of Things sensing devices are deployed in the city garbage can through embedded installation mode, including 8 HC-SR04 ultrasonic sensor arrays distributed in annular form on the upper inner wall, 4 strain gauge pressure sensors fixed on the four corners of the bottom of the can, a 32x32 pixel infrared thermal imaging sensor installed in the center of the upper part, a gas sensor group installed in the middle of the can wall, and a temperature and humidity sensor in the upper part of the can. All sensors are connected to an edge computing unit carrying an ARM Cortex-A53 processor through a waterproof level IP67 shielded cable to build a complete front-end sensing network. The pressure sensor data is collected by a 12-bit ADC every 100 milliseconds, and when the weight change exceeds 100 grams, it is marked as an incremental throwing event and triggers the ultrasonic sensor array to work. Each sensor continuously collects sound wave echo time for 5 times, and the average value after removing the maximum and minimum value is taken as the distance data. The measured distance value is converted into a spatial rectangular coordinate combined with the sound velocity value to form a garbage surface height coordinate point. The edge computing unit divides the garbage can space into a 10x10x15 three-dimensional grid, and each grid unit represents a voxel. The density distribution of each voxel is calculated by the surface height coordinate point and the weight data to generate garbage three-dimensional structure voxel data containing position and density information.

[0024] Step S2: calculating the total filling rate of garbage according to the garbage three-dimensional structure voxel data; evaluating the dynamic thermal decomposition degree data according to the garbage three-dimensional structure voxel data, and calculating the in-can state emergency degree according to the total filling rate of garbage to obtain a dynamic urgency index; performing supervision weighted clean-up urgency processing according to the dynamic urgency index, and then identifying a to-be-processed station cluster;

[0025] In the embodiment of the present application, the three-dimensional structure voxel data of garbage is obtained, the number of filled voxels is counted and divided by the total number of voxels to calculate the total filling rate of garbage. At the same time, the infrared thermal imaging sensor is called to collect the temperature distribution inside the garbage can, the temperature data space mapping is performed, and the three-dimensional temperature index voxel data is generated. The continuous area with a temperature higher than the ambient temperature by more than 5℃ is extracted as the effective heating area, and the average filling density thereof is calculated. When the density is greater than 0.8g / cm3, the high-risk organic matter weight is set to 1.5; when the density is less than 0.3g / cm3, the low-risk organic matter weight is set to 0.5. The voxels in the effective heating area are scored for activity, and the highest activity score is extracted as the dynamic thermal decomposition degree data. Then, according to the environmental humidity, a segmented function compensation is performed, and the odor diffusion influence coefficient is set in combination with the real-time gas concentration value obtained by the gas sensor and the environmental wind speed value obtained by the city meteorological data interface. Based on these parameters, the dynamic urgency index is calculated, and according to the functional area category label (commercial area, residential area or public place) of the garbage can, the monitoring weight coefficient is matched, the weighted cleaning and removal urgency is calculated and normalized, and the cleaning and removal score data is generated. The garbage cans with a cleaning and removal score greater than 75 are screened to form a high-priority site list, and the site with the highest cleaning and removal score is selected as the cluster generation center point. The sites within the range of the center point are collected to form a to-be-processed site cluster.

[0026] Step S3: extracting the to-be-visited site coordinate data in the to-be-processed site cluster, and dynamically planning the theoretical optimal collection and transportation sequence of the target collection and transportation vehicle;

[0027] In the embodiment of the present application, the geographic coordinates of all sites in the to-be-processed site cluster are extracted to construct the to-be-visited site coordinate data set. The garbage can site with the highest cleaning and removal score is set as the first target point of the target collection and transportation vehicle, the real-time driving time of the point to all remaining points is queried through a map service interface, and a candidate point time consumption table is obtained. The minimum value finding algorithm is executed to select the point with the shortest time as the next optimal work point. When the driving time of multiple points differs by no more than 60 seconds, the optimal point is determined by considering the two factors of driving time and cleaning and removal score through a comprehensive scoring mechanism. The selected work point is added to the vehicle task list, and the point is removed from the to-be-visited site coordinate data. The above path planning operation is repeatedly executed until the vehicle capacity limit or the shift time limit condition is met, and a feasible work sequence is generated. Subsequently, the key points with a cleaning and removal score significantly higher than the average value in the sequence are subjected to priority secondary weight adjustment, the point order is adjusted through a local optimization strategy, and the peak period avoidance rules of commercial area and public place category garbage cans are considered, and finally the theoretical optimal collection and transportation sequence is generated, so that the intelligent planning of garbage collection and transportation path is realized.

[0028] Step S4: processing the theoretical optimal collection and transportation sequence by segment navigation instruction, and performing single-point fine work annotation through the three-dimensional structure voxel data of garbage, so as to manage the intelligent garbage can collection and transportation work.

[0029] In the embodiments of the present application, the theoretically optimal collection and transportation sequence is extracted as an ordered latitude and longitude array, and a complete navigation path is generated through the driving route planning interface of the map open platform. The continuous path is divided into multiple independent road segments according to the stations, and each road segment constitutes a segmented navigation instruction containing the start and end point coordinates, the passing point set, the turning point set, the estimated time and distance, and other information. For each target station, analyze the garbage three-dimensional structure voxel data, identify the voxel set with a density value less than 0.3 kg / m3 and spatial continuity as the internal structural cavity, calculate the total volume of the cavity and the volume ratio of the garbage can, and evaluate the structural instability risk index. At the same time, calculate the centroid position of the garbage pile, and determine whether the horizontal offset from the can geometric center exceeds the can radius by 30% to determine the offloading state. Combine the structural instability risk and the offloading state to quantify the dumping impact force, generate a five-level estimated dynamic unloading impact value, and accordingly formulate the corresponding operation suggestions, such as regular dumping, slow dumping, segmented dumping, small-angle multiple dumping, or special handling. The segmented navigation instruction and the single-point fine operation note are integrated into a complete operation instruction package, which is pushed to the vehicle-mounted terminal of the target collection and transportation vehicle through the 4G / 5G network, realizing the fine management of the whole process of the intelligent garbage can collection and transportation operation.

[0030] Preferably, step S1 comprises the following steps:

[0031] Step S11: deploying Internet of Things sensing devices in the garbage can in the smart city, and jointly accessing an edge computing unit with at least 4 general input and output interfaces and 1 analog-to-digital conversion interface, thereby constructing an Internet of Things front-end sensing network; wherein the Internet of Things sensing devices include an ultrasonic sensor array, a pressure sensor, an infrared thermal imaging sensor, a gas sensor, and a temperature and humidity sensor;

[0032] Step S12: real-time reading of the voltage signal of the pressure sensor in the Internet of Things front-end sensing network, and obtaining the real-time total weight data of the garbage after analog-to-digital conversion;

[0033] Step S13: based on the real-time total weight data of the garbage, marking an incremental throwing event at each time when the weight changes by more than 100 grams;

[0034] Step S14: based on the time of the incremental throwing event, using the ultrasonic sensor array in the Internet of Things front-end sensing network to obtain the echo time of each sensor to the surface of the garbage, to obtain the original echo length;

[0035] Step S15: converting the original echo length into the spatial rectangular coordinates of the corresponding sensor according to the preset sound speed value, thereby determining the surface height coordinate point inside the garbage can;

[0036] Step S16: According to the surface height coordinate points and the real-time filling total weight data of the garbage, the structural voxel change analysis of the throwing event is performed to generate the garbage three-dimensional structural voxel data.

[0037] In the embodiment of the present application, when the Internet of Things sensing device is deployed inside the garbage can, an embedded installation method is adopted to fix each sensor at a preset position on the inner wall of the garbage can. The ultrasonic sensor array is composed of 8 HC-SR04 type sensors, which are evenly distributed in a ring shape on the upper inner wall of the garbage can. The pressure sensor adopts 4 strain gauge type sensors distributedly arranged and fixed at the four corners of the bottom of the garbage can. The infrared thermal imaging sensor adopts an array with a resolution of 32x32 pixels and is installed at the central position of the upper part of the garbage can. The gas sensor group includes methane, ammonia, and hydrogen sulfide detection units and is installed in the middle of the barrel wall. The temperature and humidity sensor is installed at the upper part inside the barrel. All the sensors are connected to the edge computing unit with an ARM Cortex-A53 processor through shielding cables with a waterproof level of IP67, which has 8 GPIO interfaces and 2 12-bit ADC conversion interfaces, to form a complete front-end sensing network.

[0038] During the collection process of the pressure sensor, the edge computing unit triggers a sampling interrupt every 100 milliseconds by configuring a timer to perform an analog-to-digital conversion collection operation. The analog voltage signals output by the four pressure sensors are amplified by the preamplifier circuit and then digitized by the 12-bit ADC built-in the edge computing unit, with a conversion accuracy of 0.1 g / bit. The digitized signals are filtered by a low-pass filter to remove high-frequency noise, with a cutoff frequency set to 10 Hz. The digitized values of the four pressure sensors are weighted and summed, and the weight coefficients are determined according to the stress distribution characteristics of the bottom of the garbage can, which are 0.26, 0.24, 0.25, and 0.25, respectively. The final weighted sum value is multiplied by the linear proportionality coefficient 5.78 calibrated in advance by standard weights, to obtain the accurate real-time filling total weight data of the garbage, with the unit of grams.

[0039] The real-time filling total weight data of the garbage is continuously monitored, and a sliding window method is used for data smoothing processing, with a window size of 5 sampling points. When the absolute value of the difference between two adjacent smoothed data points exceeds the threshold of 100 grams, the edge computing unit records the current timestamp in the internal time register and generates an incremental throwing event data structure containing three fields of timestamp, weight change value, and change duration. For the case where the weight suddenly increases and then quickly falls back, a dead time of 50 milliseconds is set to avoid misjudgment. After the throwing event is determined, the event information is stored in the circular buffer, and the buffer size is 256 event records, and the event index table is updated at the same time.

[0040] When the incremental throwing event is triggered, the edge computing unit immediately sends a collection command to the ultrasonic sensor array. Each ultrasonic sensor receives a 10-microsecond wide trigger pulse through the IO trigger pin, and then emits an ultrasonic signal beam with a frequency of 40 kHz. The ultrasonic signal is received by the sensor after being reflected by the surface of the garbage. The edge computing unit accurately measures the time interval from emission to reception through a hardware counter. The counter clock frequency is 1 MHz, providing a time resolution of 1 microsecond. To improve measurement accuracy, five consecutive measurements are taken for each sensor. The average of the remaining three measurements after removing the maximum and minimum values is taken as the original echo length of the sensor, with a unit of microseconds. During the measurement process, considering the multi-path reflection problem in the garbage can, the maximum detection distance is set to 1.5 meters, corresponding to a maximum echo length of 8824 microseconds.

[0041] When converting the original echo length to spatial coordinates, the sound speed is first corrected according to the current garbage can internal environment temperature. At a standard temperature of 25 degrees Celsius, the sound speed value is set to 346.3 meters per second, and the temperature correction formula is actual sound speed = 346.3 + 0.6 x (current temperature - 25). Multiply the corrected sound speed value by the original echo length and divide by 2 to get the distance value from each ultrasonic sensor to the garbage surface. According to the pre-set installation position coordinates of each sensor in the garbage can, combined with the measured distance value, the spatial rectangular coordinates of each reflection point on the garbage surface are calculated through the trigonometric function relationship. In the calculation, the center of the can bottom is taken as the coordinate origin, and the right-hand coordinate system with the vertical upward direction as the positive direction of the z-axis is used. For the area where the sensor echo is lost, the linear interpolation method of adjacent valid points is used for completion.

[0042] The garbage can space is divided into a 10x10x15 three-dimensional grid, with each grid cell representing a voxel. The voxel size is 5 cm x 5 cm x 3 cm. According to the surface height coordinate points, the occupancy state of each voxel is determined using the back projection method. Voxel below the surface is marked as occupied, and voxel above the surface is marked as unoccupied. At the same time, combined with the real-time total weight data of the garbage, the average density of the occupied voxel is calculated. For newly occurring throwing events, by comparing the voxel occupancy state changes before and after the event, the newly occupied voxel set is identified, and the density distribution of these newly occupied voxels is calculated based on the weight change value. By comparing different density thresholds (low density 0.2 g / cm3, medium density 0.6 g / cm3, high density 1.2 g / cm3), the garbage three-dimensional structure is divided into different categories, and finally the garbage three-dimensional structure voxel data containing position and density information is generated.

[0043] Preferably, the structure voxel change analysis of the throwing event in step S16 according to the surface height coordinate points and the real-time total weight data of the garbage includes:

[0044] Based on the increment throwing event, the ratio of volume increment and weight increment of each throwing event is calculated by real-time filling total weight data of surface height coordinate points and garbage, the density of different batches of garbage is evaluated, and the layered garbage density value is obtained;

[0045] The historical surface height time sequence data is acquired;

[0046] The historical surface height time sequence data and the preset three-dimensional geometric parameters of the garbage can in the edge computing unit are used to perform real-time coordinate mapping and basic voxel grid calibration, and the corresponding depth voxel is density-weighted by using the layered garbage density value, so that a density layered voxel model is generated;

[0047] The density layered voxel model is subjected to lower layer voxel grid division, and then the lower layer voxel grid is simulated and corrected for gravity sedimentation effect, that is, the layered garbage density value corresponding to the lower layer voxel grid is compressed by 1-5%, so that garbage three-dimensional structure voxel data is obtained.

[0048] In the embodiment of the application, for each increment throwing event, first, the real-time filling total weight data of garbage before and after the event is acquired, and the weight increment Δm is calculated. At the same time, based on the two groups of surface height coordinate point data before and after throwing, the volume increment Δv is calculated by a three-dimensional discrete integral method. In specific implementation, the cross section of the garbage can is divided into a 10x10 grid, and the area of each grid is 25 square centimeters. For each grid unit, the surface height difference before and after throwing is calculated, multiplied by the grid area to obtain the volume increment of the grid, and then the total volume increment Δv is obtained by summing all the grids. When calculating the density, the formula ρ=Δm / Δv is executed, where ρ represents the layered garbage density value (grams / cubic centimeter), Δm represents the weight increment (grams), and Δv represents the volume increment (cubic centimeters). The calculation result is stored in a three-dimensional array according to the current height of the garbage surface, forming a layered density record in the vertical direction, with a resolution of 3 centimeters per layer.

[0049] The edge computing unit maintains a three-dimensional ring buffer for storing historical surface height data. The size of the buffer is 10x10x24, corresponding to the 10x10 grid of the cross section of the garbage can and the time window of the last 24 hours. When a throwing event occurs, the current surface height coordinate point is mapped to the 10x10 grid to generate a height matrix, and the time stamp is attached before being stored in the ring buffer. In order to reduce the storage space occupation, a relative height storage method is adopted, that is, only the height value relative to the bottom of the barrel is recorded, with a precision of 0.5 centimeters. At the same time, for the area that has not changed for a long time (more than 30 minutes), a time marker compression algorithm is used, only the start time and end time are recorded, and data redundancy is reduced. The ring buffer is updated according to the first-in-first-out principle, and the earliest historical data is overwritten when new data is written. The surface height data at different time points is quickly retrieved through the buffer index table.

[0050] The edge computing unit pre-sets the three-dimensional geometric parameters of the garbage can, including a can bottom diameter of 50 cm, a can opening diameter of 58 cm, a can height of 90 cm, and a can wall inclination angle of 3.5 degrees. Based on these parameters, a circular truncated cone model is established, and a right-handed coordinate system is set in the edge computing unit, with the origin at the center of the can bottom and the z-axis vertically upward. The space of the garbage can is divided into a 10x10x30 basic voxel grid, with a horizontal resolution of 5 cm and a vertical resolution of 3 cm. For each voxel, the center point coordinates are calculated, and it is determined whether it is within the geometric range of the circular truncated cone. Subsequently, the historical surface height time series data is interpolated on the time axis to determine the surface morphology at each time. For voxels below the current surface, the corresponding level density value is queried according to the depth, and the density is weighted. If there is no directly measured density value at a certain depth layer, a distance weighted average algorithm is used to calculate the interpolation result by considering the density values of the four adjacent layers. Finally, a density layered voxel model containing position and density attributes is formed.

[0051] For the density layered voxel model, the voxel model is divided into upper, middle, and lower layers of grids according to the depth of the garbage layer. The upper layer grid is the voxel within the range of 0-15 cm below the surface, the middle layer grid is the voxel within the range of 15-45 cm below the surface, and the lower layer grid is the voxel within the range of 45 cm or more below the surface. For the lower layer voxel grid, the gravity settling effect simulation is performed. In specific implementation, the compression compensation coefficient k is calculated according to the garbage retention time t (unit: hour), and the calculation formula is k = 0.01 + 0.0002 x t, where t is not more than 200 hours, and k is not more than 0.05 (i.e. maximum 5%). For the density value p of each voxel in the lower layer grid, the correction calculation p corrected = p x (1 + k) is performed, where p corrected represents the density value after the gravity settling effect correction. The depth d of the voxel is also considered during the density correction, and the larger the depth, the higher the compression rate. The specific implementation is to increase the depth factor 0.0005 x (d-45) on the basis of the basic compression compensation coefficient, where d represents the depth from the surface (cm). After completing the density correction of all voxels, the position information and the corrected density value are integrated to generate garbage three-dimensional structure voxel data containing the gravity settling effect.

[0052] Especially important is that the density layered voxel model is divided into lower layer voxel grids, and then the gravity settling effect simulation correction is performed on the lower layer voxel grids, which includes:

[0053] Taking the lowest layer of the filled voxels in the density layered voxel model as the reference, all the filled voxels in this layer and above are identified as upper bearing voxels;

[0054] The weights of all upper bearing voxels are accumulated to calculate the total cumulative pressure value acting on the top surface of the lower layer garbage;

[0055] According to the total cumulative pressure value, and through a preset pressure-compression coefficient mapping table, a dynamic compression compensation coefficient between 1% and 5% is queried and determined;

[0056] The voxels located in all unfilled regions in the density stratified voxel model and under the upper load-bearing voxels are identified as the lower layer voxels to be corrected;

[0057] The stratified garbage density value corresponding to the lower layer voxels to be corrected is multiplied by the dynamic compression compensation coefficient to obtain a pressure-corrected density value, and the original model is updated using the value, so as to obtain the garbage three-dimensional structure voxel data.

[0058] In the embodiment of the application, first, the density stratified voxel model is taken as the basis, and the lowest layer position of the filled voxels is determined through a voxel state traversal algorithm. When implemented, all voxels in a 10*10 grid are scanned layer by layer from the bottom of the garbage can (z=0) upwards, and the z coordinate of the first non-zero density voxel appearing in each grid column is recorded. When at least one non-zero density voxel is detected in all grid columns, the minimum value of the z coordinates is taken as the lowest layer reference surface of the filled voxels, and the z value is recorded as zmin. Subsequently, a region growing algorithm is executed, and all voxels with a z coordinate greater than or equal to zmin and a density value greater than 0.05 g / cm3 are marked as upper load-bearing voxels. In order to improve the calculation efficiency, a three-dimensional array is used to record the state flag of each voxel, wherein a flag value of 1 indicates that the voxel is an upper load-bearing voxel. The scanning process adopts a depth-first search strategy to ensure that all connected filled voxels are correctly identified.

[0059] After determining the upper load-bearing voxels, the edge computing unit performs vertical pressure accumulation calculation. The grid columns in the vertical direction (a total of 10*10=100 grid columns) are processed one by one, and all voxels marked as upper load-bearing voxels are traversed from top to bottom for each grid column. For each load-bearing voxel, its pre-assigned density value ρi (unit: g / cm3) is read and multiplied by the voxel volume vi (unit: cm3) to calculate the mass mi of the voxel, where vi=5*5*3=75 cm3. For grid column j, the total mass Mj of all upper load-bearing voxels in the column is calculated, where i represents the serial number of all upper load-bearing voxels in the grid column. Then the mass is converted into force, Fj=Mj*9.8*10 ―3 , where Fj is the pressure generated by grid column j (unit: Newton), 9.8 is the acceleration of gravity (unit: m / s2), and 10^(-3) is the unit conversion factor. Finally, the pressures of all grid columns are summed up to obtain the total cumulative pressure value Ftotal=∑Fj acting on the top surface of the lower layer garbage, where j represents the grid column number (1 to 100).

[0060] The edge computing unit is pre-installed with a pressure-compression coefficient mapping table, which is obtained through material mechanics testing. The testing process uses a standard garbage sample (containing typical urban garbage components such as plastic, paper, and organic matter, mixed in a ratio of 4:3:3) to obtain compression rate data under different pressure conditions. The mapping table contains 11 discrete data points: [0 Newton, 1.0%], [50 Newton, 1.4%], [100 Newton, 1.8%], [150 Newton, 2.2%], [200 Newton, 2.6%], [250 Newton, 3.0%], [300 Newton, 3.4%], [350 Newton, 3.8%], [400 Newton, 4.2%], [450 Newton, 4.6%], [500 Newton and above, 5.0%]. According to the total cumulative pressure value Ftotal, a binary search is performed in the mapping table to locate the two closest discrete points. If Ftotal falls between the two discrete points, the corresponding dynamic compression compensation coefficient k is calculated by linear interpolation, with the interpolation formula being k = k1 + (k2 - k1) x (Ftotal - F1) / (F2 - F1), where F1 and F2 are the pressure values of the adjacent discrete points, and k1 and k2 are the corresponding compression coefficients. The calculation result is rounded to three decimal places.

[0061] The lower layer voxels to be corrected are identified by performing a three-dimensional scanning algorithm. First, a Boolean three-dimensional array of the same size as the voxel model is generated, with all initial values set to false. For each upper load-bearing voxel, record its x-y coordinate position on the horizontal plane, and mark the Boolean value of all voxel positions below this coordinate position with a z coordinate less than the z coordinate of the load-bearing voxel as true. At the same time, check the density value in the original density layered voxel model at this position, and only keep the true value mark when the density value is zero (indicating an unfilled area), otherwise reset it to false. Through horizontal plane projection, all voxels located directly below each upper load-bearing voxel in each grid column and currently unfilled are identified. To avoid false corrections, a connectivity check is also performed to ensure that there is a connected path between the voxels to be corrected and the lowest layer reference surface, excluding hollow areas blocked by other objects.

[0062] According to the position set of the lower layer voxel to be corrected and the dynamic compression compensation coefficient k, the density layered voxel model is updated. First, a temporary three-dimensional array is created for storing the corrected density value. For each lower layer voxel to be corrected, retrieve its nearest non-zero density bearing voxel directly above, and obtain the layered garbage density value p0 of the bearing voxel. Then calculate the pressure corrected density value p1 = p0 x (1+k / 100), where k is the dynamic compression compensation coefficient in percentage form. For example, when k = 3.5%, p1 = p0 x 1.035. To simulate the change of physical properties after compression, a secondary correction is also made according to the depth difference. The farther away from the bearing voxel, the smaller the correction coefficient, and the correction formula is p2 = p1 x (1-0.02 x Δz / 30), where Δz is the z coordinate difference (unit: centimeter) between the voxel to be corrected and its corresponding bearing voxel, and 30 is the normalization coefficient. The calculated p2 is the final corrected density value, which is written into the corresponding position of the temporary array. After completing the processing of all the lower layer voxels to be corrected, the temporary array is combined with the original density layered voxel model to obtain the complete garbage three-dimensional structure voxel data containing the effect of gravity settling.

[0063] Preferably, the step S2 of evaluating the dynamic thermal decomposition degree data according to the garbage three-dimensional structure voxel data and calculating the barrel status urgency according to the total garbage filling rate comprises:

[0064] The temperature distribution information inside the garbage can is collected by the infrared thermal imaging sensor in the Internet of Things front-end sensing network to obtain the barrel infrared temperature data;

[0065] The garbage three-dimensional structure voxel data is used as a spatial reference to perform pixel-level registration and superposition according to the barrel infrared temperature data to generate three-dimensional temperature index voxel data;

[0066] The ambient temperature and ambient humidity are obtained by the temperature and humidity sensor in the Internet of Things front-end sensing network;

[0067] The dynamic thermal decomposition degree data is evaluated according to the three-dimensional temperature index voxel data and the ambient temperature;

[0068] The dynamic thermal decomposition degree data is compensated for environmental risk based on the ambient humidity, and the barrel status urgency is calculated according to the total garbage filling rate to obtain the dynamic urgency index.

[0069] In the embodiment of the present application, the infrared thermal imaging sensor in the Internet of Things front-end sensing network adopts a 32x32 pixel array structure, and the sampling frequency is set to once every 10 minutes. During the collection process, the edge computing unit sends a collection instruction to the infrared thermal imaging sensor through a serial peripheral interface, triggering the sensor to perform a complete scan. The sensor uses a non-cooled microbolometer array, with a temperature measurement range of -20°C to 120°C and a temperature resolution of 0.1°C. The sensor has a built-in 16-bit ADC conversion circuit that converts thermal radiation signals into digital temperature values. Since the sensor is installed at the central position of the upper part of the garbage can, a 120° wide-angle lens is used to cover the entire space inside the can, and the raw data obtained is a planar thermal image. After the edge computing unit receives the 32x32 pixel temperature matrix, it performs Gaussian filtering to eliminate thermal noise, with a filter kernel size of 3x3 and a standard deviation of 0.8.

[0070] The obtained garbage three-dimensional structure voxel data is used as a spatial reference benchmark to perform temperature data spatial mapping. First, a geometric transformation matrix of the infrared thermal imaging sensor and the garbage can three-dimensional coordinate system is established, and the transformation matrix is determined by the correspondence of the five feature points calibrated in advance, with the calibration points located at the center top and four edge tops of the garbage can. For each pixel point in the 32x32 temperature matrix, the ray equation in the three-dimensional space is determined by the back projection method, and then the intersection with the surface voxel in the garbage three-dimensional structure voxel data is calculated to obtain the three-dimensional space coordinates of the temperature point. For voxels that cannot be directly observed (such as parts obscured by other garbage), a temperature decay model is used for estimation, with a decay coefficient of 0.15 / cm. During registration, the Iterative Closest Point (ICP) algorithm is used to eliminate the deviation caused by sensor installation errors, and the registration accuracy is controlled within 2 cm. After completing the spatial mapping, each voxel is assigned a corresponding temperature attribute value in addition to the density information, forming three-dimensional temperature voxel data containing spatial position, density, and temperature attributes.

[0071] The temperature and humidity sensor in the Internet of Things front-end sensing network adopts a digital composite sensor, with a temperature measurement range of -40°C to 80°C and an accuracy of ±0.5°C; the humidity measurement range is 0% to 100% relative humidity, with an accuracy of ±3%. The sensor is fixedly installed on the inner wall of the upper part of the garbage can, 5 cm away from the can opening, avoiding direct contact with garbage to ensure the measurement of air parameters. The edge computing unit sends a collection instruction to the temperature and humidity sensor through a single bus interface every 5 minutes, and the sensor returns a 16-bit digital signal, with the high 8 bits being humidity data and the low 8 bits being temperature data. The received data is checked for validity. To eliminate the influence of instantaneous fluctuations, median filtering is applied to the continuous 3 times of collected data to obtain stable environmental temperature and humidity readings. For temperature and humidity outliers (temperature <-10°C or >50°C, humidity <10% or >95%), a self-checking program is started and re-collection is performed, and if the continuous 3 times of abnormalities are marked as sensor fault state.

[0072] The degree of waste thermal decomposition activity is evaluated based on three-dimensional temperature scale voxel data. First, the difference ΔT = Tv - Te between the temperature of the waste voxel Tv and the ambient temperature Te is calculated, where Tv represents the voxel temperature (°C) and Te represents the ambient temperature (°C). According to the ΔT value, the waste voxel is divided into four categories: inert zone (ΔT < 2°C), low activity zone (2°C ≤ ΔT < 5°C), medium activity zone (5°C ≤ ΔT < 10°C), and high activity zone (ΔT ≥ 10°C). The proportion of the number of voxels in each activity zone to the total filled voxels is calculated to obtain three proportion coefficients: low activity proportion Pl, medium activity proportion Pm, and high activity proportion Ph. Hot spot clustering analysis is performed, and the DBSCAN algorithm is used to identify temperature anomaly aggregation areas, with clustering parameters set as: neighborhood radius 12 cm, minimum point number 5, and the identified cluster center marked as a hot spot. For each hot spot, the difference ΔTmax between its maximum temperature Tmax and the ambient temperature is calculated, and the hot spot volume Vh is recorded. Finally, based on the proportion coefficient and hot spot parameters, the dynamic thermal decomposition degree H = 0.2 × Pl + 0.5 × Pm + 1.0 × Ph + 0.05 × n × (ΔTmax / 10) × (Vh / 100) is calculated, where n is the number of hot spots.

[0073] The dynamic thermal decomposition degree data is risk-compensated based on environmental humidity. First, the humidity risk factor R = 1 + [(RH - 50) / 100] 2 is calculated, where RH represents the ambient relative humidity percentage value. When RH < 50%, dry environment inhibits decomposition but increases fire risk; when RH > 50%, humid environment accelerates organic matter decomposition and gas production. Subsequently, the total filling rate of the waste F = Vf / Vt is calculated, where Vf is the total volume of filled voxels (cubic centimeters) and Vt is the total volume of the garbage can (cubic centimeters). The F value is divided into four intervals corresponding to different basic urgency coefficients Cf: mild filling (F < 30%) corresponds to Cf = 0.6, moderate filling (30% ≤ F < 60%) corresponds to Cf = 1.0, severe filling (60% ≤ F < 85%) corresponds to Cf = 1.5, and critical filling (F ≥ 85%) corresponds to Cf = 2.5. Comprehensive calculation is performed to obtain the dynamic urgency index U = H × R × Cf, where H is the dynamic thermal decomposition degree, R is the humidity risk factor, and Cf is the filling urgency coefficient. When the calculation result U exceeds the warning threshold value 12, the garbage can state abnormal alarm is triggered, and the scheduling center decision is executed through the smart city Internet of Things device data collection and management platform.

[0074] Preferably, evaluating the dynamic thermal decomposition degree data based on three-dimensional temperature scale voxel data and ambient temperature includes:

[0075] Extracting a continuous area with a temperature higher than the ambient temperature by more than 5°C from the three-dimensional temperature scale voxel data as an effective heating area;

[0076] The average filling density of the garbage in the effective heat generation area is calculated through the three-dimensional structure voxel data of the garbage;

[0077] When the average filling density in the effective heat generation area is greater than a preset first content threshold, the content risk weight factor of the effective heat generation area in the garbage can is set to 1.5, which is recorded as a high-risk organic matter weight;

[0078] When the average filling density in the effective heat generation area is less than a preset second content threshold, the content risk weight factor of the effective heat generation area in the garbage can is set to 0.5, which is recorded as a low-risk organic matter weight; wherein the preset first content threshold is greater than the preset second content threshold;

[0079] According to the high-risk organic matter weight and the low-risk organic matter weight, the voxel activity score of the effective heat generation area is processed, and then the highest activity score in the voxel is extracted as the dynamic heat decomposition degree data of the decomposition activity degree of the garbage.

[0080] In the embodiment of the application, first, temperature abnormal voxels are extracted from the three-dimensional temperature scale voxel data. The measured ambient temperature Te is read, and the temperature screening threshold is set to Te+5℃. By comparing the temperature value Tv of each voxel with the size relationship of the screening threshold, the voxels satisfying the condition Tv>Te+5℃ are marked as thermal abnormal voxels. Subsequently, three-dimensional connected domain analysis is performed, and all thermal abnormal voxels are clustered using the 26-connectivity criterion (considering the face, edge and vertex adjacency of the voxel). The connected domain analysis uses a depth-first search algorithm, which starts from an unmarked thermal abnormal voxel, recursively searches all thermal abnormal voxels adjacent to it and marks them as the same region number. When the search is completed, the connected domain with a voxel number greater than 5 is screened as an effective heat generation area, and each effective heat generation area is assigned a unique region identifier. The three-dimensional coordinate range of the effective heat generation area and the voxel list contained are stored in the region attribute table, and the average temperature, maximum temperature and volume information of each region are recorded.

[0081] The effective heat-emitting regions are spatially registered with the 3D structure voxel data of the garbage by a region matching algorithm. First, a spatial index table is established to organize each voxel in the 3D structure voxel data of the garbage in a 10x10x30 3D grid structure for fast query. For each effective heat-emitting region R, all the voxel positions contained therein are traversed, and the density values ρi (unit: g / cm3) at the corresponding positions are extracted from the 3D structure voxel data of the garbage. When a position does not have a density record in the original data, an estimated value is obtained from adjacent voxels by using a nearest neighbor interpolation method. An arithmetic average calculation is performed on all the extracted density values to obtain the average filling density ρavg,R of the region, where nR represents the number of voxels contained in the region R. Extreme outliers (density values less than 0.05 g / cm3 or greater than 2.0 g / cm3) are removed during the calculation to reduce the influence of measurement errors. For a garbage bin containing multiple effective heat-emitting regions, the average filling density value is calculated and recorded for each region separately and stored in the density field of the region attribute table.

[0082] The first content threshold is preset to be 0.8 g / cm3, which is obtained by laboratory tests. The test process uses the volume density data of representative organic garbage samples (including food residue, fruit peel, vegetable leaves, and other wet garbage) under different compaction degrees to determine. When the average filling density ρavg,R of the effective heat-emitting region is greater than 0.8 g / cm3, a high-risk organic matter determination process is performed. First, the region is marked as a high-density organic area, and the region type flag in the region attribute table is updated. Subsequently, the content risk weight factor WR of the region is set to 1.5, which is recorded as the high-risk organic matter weight. At the same time, the monitoring frequency of the region is increased, and the temperature sampling interval is shortened from the standard 10 minutes to 5 minutes to more closely monitor the temperature change trend of the high-risk region.

[0083] The second content threshold is preset to be 0.3 g / cm3, which is determined by density measurement of dry garbage (such as paper, plastic, fabric, etc.). It is ensured that the first content threshold 0.8 g / cm3 is greater than the second content threshold 0.3 g / cm3, forming a three-section density classification interval. When the average filling density ρavg,R of the effective heat-emitting region is less than 0.3 g / cm3, a low-risk organic matter determination process is performed. The region is marked as a low-density region, and the region type flag in the region attribute table is updated to 1. The content risk weight factor WR of the region is set to 0.5, which is recorded as the low-risk organic matter weight. For regions with an average filling density between the second content threshold and the first content threshold (i.e., 0.3 g / cm3≤ ρavg,R≤ 0.8 g / cm3), the content risk weight factor is set to 1.0, which is recorded as a medium-risk region. The region type flag in the region attribute table is updated to 2, and the temperature sampling interval is maintained at the standard 10 minutes.

[0084] The voxel activity score processing is performed for all the valid heat-emitting regions. For each voxel i in a valid heat-emitting region R, the temperature difference ΔΤί = Τί - Τε is calculated, where Τί is the voxel temperature and Τε is the ambient temperature. Subsequently, the base activity score Si,base = 0.1 x ΔΤί is calculated based on the temperature difference, and the base activity score is capped at 2.0 when ΔΤί ≥ 20°C. Further considering the temperature gradient factor, the maximum temperature difference ΔΤί, max of the voxel and its six neighboring voxels is calculated, and the gradient activity component Si,grad = 0.05 x ΔΤί, max is added. The integrated activity score Si = WR x (Si,base + Si,grad) of the voxel is calculated based on the risk weight factor WR. The maximum value Smax = max{Si} of the integrated activity scores of all the voxels in all the valid heat-emitting regions is found as the dynamic thermal decomposition degree data of the garbage decomposition activity. When there are multiple valid heat-emitting regions in the garbage bin, the global maximum activity score is taken as the final result. The value usually ranges from 0 to 5, and the higher the value, the more intense the garbage thermal decomposition activity and the more active the decomposition process.

[0085] In one implementation of the embodiments of the present application, please refer to Fig. 2 , which is a voxel activity score graph. For example, in the actual application of a smart garbage disposal station, three-dimensional temperature scale voxel data is collected, and the ambient temperature Te is measured as 25°C. As Fig. 2 , the temperature screening threshold is set as Te + 5°C = 30°C, and the voxel temperatures are screened one by one. In a typical garbage bin monitoring case, 1847 abnormal voxels with a temperature range of 30°C-55°C are detected. The three-dimensional connected component analysis is performed using the 26-connectivity criterion, and 23 connected components are identified by the depth-first search algorithm. After the voxel quantity screening (> 5 voxels), 8 valid heat-emitting regions are finally determined, which are marked as R1-R8. Among them, the largest heat-emitting region R1 contains 312 voxels, and the temperature distribution is between 32°C-48°C, with an average temperature of 39.5°C. The spatial registration algorithm is used to match the valid heat-emitting regions with the garbage three-dimensional structure voxel data. Taking the heat-emitting region R1 as an example, the density data at the corresponding position is extracted, and 2 extreme abnormal values (density < 0.05 g / cm 3 ) are removed. The average filling density ρavg, R1 = 0.85 g / cm 3 is calculated. As shown in the risk grading standard in Fig. 2 , since ρavg, R1 = 0.85 g / cm 3 > 0.8 g / cm 3(the first content threshold), the system determines the R1 region as a high-risk organic matter region, and sets the risk weight factor WR1 = 1.5. At the same time, the average filling density of the region R3 is 0.25 g / cm 3 <0.3g / cm 3 (the second content threshold), which is determined as a low-risk region, and the weight factor WR3 = 0.5. The density of the region R2 is 0.6 g / cm 3 , between the two thresholds, is classified as a medium-risk region, and the weight factor WR2 = 1.0. The voxel activity score processing is performed on each effective heat generation region. Taking the voxel with the highest temperature in the R1 region as an example, its temperature Ti = 48℃, and the temperature difference ΔTi = 48-25 = 23℃. Since ΔTi > 20℃, the basic activity score reaches the ceiling value Si,base = 2.0. Further, the temperature gradient factor is calculated, and the maximum temperature difference ΔTi,max = 12℃ between this voxel and the adjacent 6 face voxels, and the gradient activity component is 0.05x23 = 1.15. Combined with the high-risk weight factor WR1 = 1.5, the comprehensive activity score of this voxel is: Si = (2.0 + 1.15) x 1.5 = 4.73, which is highly consistent with the activity score of about 4.7 corresponding to the temperature point of 48℃ in the high-risk weight curve (red line) in Fig. 2 . By traversing all the voxels of the 8 effective heat generation regions, the highest activity scores of the regions are calculated as: R1 = 4.73, R2 = 3.12, R3 = 1.85, R4 = 2.96, R5 = 3.45, R6 = 2.21, R7 = 4.15, and R8 = 3.78. Through comparison of the monitoring data for 7 consecutive days, the accuracy rate of the garbage can thermal decomposition degree evaluation using the method of the present application reaches 92.3%, which is improved by 35% compared with the traditional single-point temperature monitoring method. As shown in Fig. 2 , the multi-weight factor scoring mechanism effectively distinguishes the decomposition activity degree of different types of garbage.

[0086] Preferably, the dynamic thermal decomposition degree data is compensated for environmental risk based on environmental humidity, and the in-can status urgency is calculated according to the total garbage filling rate, including:

[0087] The dynamic thermal decomposition degree data is gain-compensated according to the environmental humidity to generate compensated thermal decomposition degree data;

[0088] The real-time gas concentration value is obtained by using the gas sensor in the Internet of Things front-end sensing network;

[0089] The local meteorological data is obtained, and the current environmental wind speed is extracted as the local environmental wind speed value;

[0090] The local environmental humidity is obtained, and the current environmental humidity is extracted as the local environmental humidity value;determine whether the real-time gas concentration value is greater than 50 ppm and the local environmental wind speed value is less than 1 m / s, if yes, set the influence coefficient to 1.4, if not, set to 1.0, obtain the odor dispersion influence coefficient;

[0091] The compensation thermal decomposition degree data is processed by the odor dispersion influence coefficient, and the in-barrel state emergency degree is calculated according to the total filling rate of the garbage, so as to obtain the dynamic urgency index.

[0092] In the embodiment of the application, the environmental humidity value RH (relative humidity percentage) is obtained. The segmented function compensation model is executed according to different humidity intervals to process the dynamic thermal decomposition degree data H. When the environmental relative humidity is less than 30%, the low humidity compensation coefficient KL=0.85 is adopted to inhibit the thermal decomposition activity; when the environmental relative humidity is in the interval of 30% to 70%, the linear transition compensation coefficient KM=0.85+0.015×(RH-30) is adopted; when the environmental relative humidity is higher than 70%, the high humidity compensation coefficient KH=1.45+0.01×(RH-70) is adopted to accelerate the thermal decomposition activity. The compensation calculation formula is HC=H×K, wherein HC is the compensation thermal decomposition degree data, H is the original dynamic thermal decomposition degree data, and K is the compensation coefficient corresponding to the humidity interval. When the environmental relative humidity exceeds 85% and the original thermal decomposition degree data is greater than 2.5, an additional anaerobic fermentation risk correction is performed, that is, a fixed compensation value of 0.8 is added to the compensation result.

[0093] The gas sensor group in the Internet of Things front-end sensing network includes three independent sensing units: a methane sensor (model MQ-4), an ammonia sensor (model MQ-137), and a hydrogen sulfide sensor (model MQ-136). The sensor group is connected to the edge computing unit through a dedicated acquisition circuit, and the sampling frequency is set to once every 2 minutes. During acquisition, the edge computing unit selects the three sensor channels in turn through a multiplexer, and performs 16-bit ADC conversion on the analog voltage signal output by each sensor. Five data of each sensor are collected, and the average value is taken after removing the highest and lowest values to eliminate the influence of instantaneous fluctuations. The digital signal obtained by conversion is converted into the actual gas concentration through the pre-calibrated sensor response curve. The conversion formula is C=a×(V / V0) b , wherein C is the gas concentration (ppm), V is the current ADC reading, V0 is the ADC reading under clean air, and a and b are sensor calibration coefficients. For the concentration values of methane, ammonia, and hydrogen sulfide, a weighted average method is used to calculate a comprehensive gas index, and the weights are 0.3, 0.4, and 0.3, respectively, to obtain the real-time gas concentration value G, which is in ppm. When any single gas concentration exceeds the preset alarm threshold (methane 10000 ppm, ammonia 100 ppm, hydrogen sulfide 50 ppm), a gas over-limit alarm is triggered immediately.

[0094] Local weather information is obtained through the weather data interface of the urban IoT data center. The MQTT protocol is used to subscribe to the weather data packets published by the urban weather service at every whole hour. The weather data packet contains temperature, humidity, air pressure, wind direction, wind speed, and other parameters, and the data format is JSON structure. The edge computing unit parses the received data packet and extracts the wind speed field value as the local environmental wind speed reference. Due to the distance difference between the urban weather station and the actual location of the garbage can, the location is corrected through the geographic information mapping table. The correction formula is WS = WSbase x Karea, where WS is the corrected local environmental wind speed value (m / s), WSbase is the wind speed reference value measured by the weather station (m / s), and Karea is the regional terrain correction coefficient. The regional terrain correction coefficient is obtained by analyzing the wind speed attenuation characteristics of different regions of the city in advance and stored in the regional parameter table of the edge computing unit. When the weather data interface has not been updated for more than 15 minutes, the last valid data is used and a 10% uncertainty correction is added. When the local environmental wind speed value is less than 0.2 m / s, it is determined to be a windless state, and the wind speed value is fixed at 0.2 m / s.

[0095] The real-time gas concentration value G is compared with the threshold value of 50 ppm, and the local environmental wind speed value WS is compared with the threshold value of 1 m / s. The threshold value of 50 ppm is determined based on the human olfactory perception threshold and the public area odor control standard, while the threshold value of 1 m / s is determined based on the critical wind speed value in the outdoor gas diffusion dynamics model. Four branches of judgment logic are designed: when G > 50 ppm and WS < 1 m / s, the conditions of high concentration gas and low wind speed are met, and the odor diffusion influence coefficient ID is set to 1.4; when G > 50 ppm but WS ≥ 1 m / s, the high concentration gas can be effectively diffused under sufficient wind speed, and ID is set to 1.2; when G ≤ 50 ppm but WS < 1 m / s, although the gas concentration is low, the diffusion is limited, and ID is set to 1.1; when G ≤ 50 ppm and WS ≥ 1 m / s, the low concentration gas is fully diffused under sufficient wind speed, and ID is set to 1.0.

[0096] The compensation thermal decomposition degree data HC is processed by the public health sensitivity of the odor diffusion influence coefficient ID, and the calculation formula is HS=HC×ID, wherein HS is the thermal decomposition data after sensitivity processing. Subsequently, the total filling rate F of the garbage is calculated. The filling rate F=Nfilled / Ntotal×100 is obtained by dividing the number of filled voxels Nfilled in the garbage three-dimensional structure voxel data by the total number of voxels Ntotal in the garbage can space. The filling emergency coefficient CF is determined according to the filling rate F. When F<40%, CF=0.8; when 40%≤F<65%, CF=1.0+0.01×(F-40); when 65%≤F<85%, CF=1.25+0.0375×(F-65); and when F≥85%, CF=2.0. The dynamic urgency index is obtained by performing comprehensive calculation. When U>5, the garbage can state is marked as needing attention; when U>8, it is marked as needing to be cleaned as soon as possible; and when U>12, it is marked as urgent cleaning. The dynamic urgency index U and the corresponding state mark are uploaded in real time to the smart city Internet of Things equipment data collection and management platform through a wireless network.

[0097] Preferably, the supervision weighted cleaning urgency processing according to the dynamic urgency index in step S3 comprises:

[0098] obtaining a functional area category label of the garbage can; matching a supervision weight coefficient according to the functional area category label; wherein the functional area category label comprises a commercial area, a residential area, and a public place;

[0099] performing weighted cleaning urgency calculation according to the supervision weight coefficient and the dynamic urgency index, and performing priority score normalization processing to generate cleaning score data;

[0100] setting a collection and transportation starting threshold value as 75, screening garbage cans with cleaning score data greater than 75 to obtain a high-priority site list; otherwise, a sub-priority site list is obtained;

[0101] selecting a site with the highest cleaning score from the high-priority site list as an initial center of regional clustering, as a cluster generation center point;

[0102] taking the cluster generation center point as the center, a range with a radius of 1 kilometer is drawn on the map, and all high-priority sites and part of sub-priority sites falling within the range are collected to obtain a potential cluster member set;

[0103] calculating the estimated total volume and total weight of cleaning according to the garbage three-dimensional structure voxel data to obtain a cluster estimated load;

[0104] matching a target collection and transportation vehicle closest to the cluster generation center point;

[0105] The cluster estimated load is compared with the loadable weight of the target collection vehicle. If it exceeds, the site with the lowest urgency is removed from the potential cluster member set until the condition is met, and a set of sites to be processed is generated.

[0106] In the embodiment of the present application, an electronic map database of city functional areas is established in the smart city Internet of Things management platform, containing detailed regional functional attribute information, with latitude and longitude coordinates as indexes. According to the position coordinates (accuracy ± 3 meters) uploaded by the built-in GPS positioning module of the garbage can, the corresponding functional area category label is extracted through a spatial query matching algorithm. The functional area category label is divided into three categories: commercial area (including shopping mall, office building, catering street, etc.), residential area (including residential area, apartment, community center, etc.), and public place (including park, square, station, school, etc.). According to the flow density, garbage generation characteristics and supervision requirements of different functional areas, the corresponding supervision weight coefficient is set: the weight coefficient of commercial area is 1.4, the weight coefficient of residential area is 1.0, and the weight coefficient of public place is 1.2. These weight coefficients are based on the statistical analysis of the cleaning and transportation records of the city sanitation department in the past three years and the distribution of complaints.

[0107] Based on the supervision weight coefficient W and the dynamic urgency index U, the weighted cleaning and transportation urgency is calculated. The calculation formula is P = U × W × 10, where P is the original cleaning and transportation urgency score, U is the dynamic urgency index, W is the supervision weight coefficient, and 10 is the basic score adjustment factor. Then, the priority score normalization processing is performed, using the Min-Max standardization method, the formula is S = (P-Pmin) / (Pmax-Pmin) × 100, where S is the normalized cleaning and transportation score data, P is the original cleaning and transportation urgency score, Pmin is the minimum original cleaning and transportation urgency score of all garbage cans, and Pmax is the maximum value. When the number of garbage cans is less than 10, Pmin is fixed at 0 and Pmax is fixed at 150 to avoid small sample normalization fluctuations. To ensure the real-time effectiveness of the cleaning and transportation score data, the garbage cans in the network are recalculated and sorted every 30 minutes to generate updated cleaning and transportation score data. The calculation result retains the integer part, with a value range of 0 to 100.

[0108] A collection and transportation initiation threshold of 75 points is set. This threshold is determined based on the service standards set by the city's sanitation department, indicating that collection should be prioritized when the collection score reaches or exceeds this value. A data filtering operation is performed, comparing the collection score data of all trash cans with the threshold of 75. For trash cans with a collection score greater than 75, their equipment identification number, geographic coordinates, collection score data, and functional area category label are organized into structured data entries and added to the high-priority site list. For trash cans with a collection score less than or equal to 75, data entries of the same format are added to the second-priority site list. Both lists are sorted in descending order of collection score data for easier subsequent processing. A 20-minute refresh cycle is set for the high-priority site list. When the high-priority site list is empty, the collection and transportation initiation threshold is temporarily lowered to 65, and the filtering operation is re-executed to ensure continuous collection operations.

[0109] From the list of high-priority sites, extract the site with the highest waste disposal score using a heap sort algorithm. Read the site's geographical coordinates (longitude λ, latitude λ). This is set as the initial center for regional clustering, i.e., the cluster generation center point. When multiple sites have the same collection score (score difference less than 0.5), a secondary sorting is performed, prioritizing sites with the functional area category label of commercial area, followed by public places, and lastly residential areas. If multiple sites are still tied, the site closest to the nearest collection vehicle is selected as the center point. A cluster identifier is created for the selected cluster generation center point, in the format of area code-date-serial number, such as HD-0718-01. The relevant information of the cluster generation center point (including site identifier, geographical coordinates, collection score, and functional area category label) is stored in the center point record table of the cluster management database, and the status is marked as active.

[0110] A spatial range query is performed with the cluster generation center point as the center. Geographical distance is calculated using latitude and longitude coordinates, employing the Havesing formula. Calculate the distance between two points, where d is the distance (in kilometers) and r is the Earth's radius (6371 kilometers). λ1 represents the latitude and longitude (in radians) of the center point. λ2 is the latitude and longitude (radian) of the point to be detected. All garbage can sites with a distance less than or equal to 1 km are screened, and the sites in the high-priority site list that meet the conditions are all included in the potential cluster member set. For sites in the low-priority site list, a multi-level screening standard is applied: the low-priority site clean-up score needs to be greater than 65 within 0.5 km of the center point; the low-priority site clean-up score needs to be greater than 70 within 0.5-0.8 km of the center point; the low-priority site is only included within 0.8-1 km of the center point when the clean-up score is greater than 72 and the functional area category label is the same as that of the center point. All sites that meet the conditions are organized into a potential cluster member set, and the site identification number, geographic location coordinates, clean-up score, functional area category label, and distance from the center point of each member are recorded.

[0111] The garbage three-dimensional voxel data of each garbage can in the potential cluster member set is retrieved from the database. The voxel data of each garbage can is processed in turn, and its estimated clean-up volume and weight are calculated. For volume calculation, the number of voxels with non-zero density in each garbage can n is counted, multiplied by the volume of a single voxel v (75 cubic centimeters), to obtain the garbage volume V = n x v. For weight calculation, the density value ρi of each non-zero voxel is multiplied by the voxel volume v to obtain the garbage weight m = ∑(ρi x v). Considering the garbage compression factor, a compression correction coefficient of 0.85 is applied to the volume calculation to obtain the compressed volume V' = V x 0.85. The volume and weight data of all garbage cans in the potential cluster member set are summarized, and the total volume Vtotal = ∑V' and the total weight mtotal = ∑m are calculated. Based on these two indicators, the cluster estimated load capacity is represented in the format "total weight: total volume", for example, "350 kg: 0.8 m3".

[0112] A real-time updated collection and transportation vehicle state database is maintained, recording the vehicle number, vehicle model specification, maximum load capacity, maximum loading volume, current position coordinates, and working status of each vehicle. Taking the cluster generation center point as the reference point, the distances of all idle collection and transportation vehicles from the point are calculated. The distance calculation also uses the Haversine formula, and the result is accurate to meters. All idle vehicles are arranged in ascending order of distance, with the closest vehicle being given priority. If there are multiple idle vehicles within the closest distance (within 2 km), a multi-factor scoring is performed: vehicle distance score = 100 - distance (km) x 20, load matching degree score = 100 - |vehicle maximum load - estimated total weight| / estimated total weight x 100 (upper limit is 100), total score = 0.6 x distance score + 0.4 x load matching degree score. The vehicle with the highest total score is selected as the target collection and transportation vehicle. When all vehicles are in a busy state, the vehicle that is expected to complete the current task earliest is queried, and its estimated idle time and the urgency of the current cluster are calculated. If the waiting time exceeds 30 minutes and the cluster contains a site with a score exceeding 90, the emergency scheduling mechanism is triggered.

[0113] The maximum load mmax (kg) and the maximum loading volume Vmax (m3) of the target collection and transportation vehicle are read from the vehicle database. The total weight mtotai and the total volume Vtotai in the cluster estimated load are compared with the vehicle parameters. When mtotai > mmax or Vtotai > Vmax, the cluster member adjustment operation is performed. First, all sites are arranged in ascending order of the collection and transportation score from the potential cluster member set, and then sites are removed from the end one by one, while the total weight and total volume values are decremented. After each site is removed, mtotai and Vtotai of the remaining sites are recalculated and compared with mmax and Vmax until the conditions of mtotai≤mmax and Vtotai≤Vmax are met. To ensure the reasonableness of the number of sites in the cluster, the minimum number of retained sites is set to 3 and the maximum number of retained sites is set to 15. If the number of sites that meet the load conditions exceeds 15, only the top 15 sites with the highest scores are retained; if the load conditions are still not met after removing low-score sites and the number of remaining sites has reached 3, the target collection and transportation vehicle is replaced with a vehicle model with a larger load. The finally determined site set constitutes a to-be-processed site cluster, a collection and transportation task sheet is generated, containing detailed information of all sites in the cluster, and the task is assigned to the target collection and transportation vehicle.

[0114] Preferably, step S3 comprises the following steps:

[0115] Step S31: The highest-priority garbage can in the to-be-visited site coordinate data is taken as the first target point of the target collection and transportation vehicle, and then the real-time driving time of all remaining point positions in the to-be-visited site coordinate data is queried through a preset map service interface, to obtain candidate point position time consumption;

[0116] Step S32: The point position with the shortest time consumption is selected from the candidate point position time consumption to determine the next optimal working point of the vehicle;

[0117] Step S33: The next optimal working point is added to the task list of the current vehicle, and the point is removed from the to-be-visited site coordinate data, to obtain an updated working path;

[0118] Step S34: Steps S31 to S33 are repeated until the capacity or shift time limit of the target collection and transportation vehicle is met, and the updated working path is taken as a feasible working sequence;

[0119] Step S35: The feasible working sequence is subjected to secondary priority weight adjustment, to obtain a theoretically optimal collection and transportation sequence.

[0120] In the embodiment of the application, the garbage can site with the highest collection and transportation score data is extracted from the to-be-processed site cluster, and the geographic coordinates (longitude λ1, latitude φ1) of the site are taken as the first target point of the target collection and transportation vehicle. ) is specified as the first target point of the target collection and transportation vehicle. The preset Gaode map service interface is called through the RESTful API method to construct a batch path planning request. The request parameters include: the starting point coordinates (the longitude and latitude coordinates of the first target point), the terminal point coordinate set (the longitude and latitude coordinates of all remaining points in the coordinate data of the to-be-visited station), the driving strategy (set to the shortest time), and the vehicle type (set to a heavy truck according to the model of the target collection and transportation vehicle). An HTTP POST request is sent, carrying a JSON data body in a specified format, and the request timeout time is set to 3 seconds. The map service interface returns JSON format data containing all path schemes, and the expected driving time (unit: seconds), distance (unit: meters) and real-time traffic information of each path are extracted through a JSON parser. When the interface call fails, a backup offline path calculation module is started to estimate the driving time based on the Euclidean distance and the historical average driving speed (18 kilometers / hour).

[0121] A minimum value finding algorithm is performed on the candidate point time consumption table. First, a hash table structure is established with the garbage can station identifier as the key and the driving time as the value, and then a linear scan is performed to record the index of the point with the minimum driving time. When multiple point driving times are found to differ by no more than 60 seconds, a multi-factor evaluation mechanism is triggered: the comprehensive score S = 0.7 x (1-T / Tmax) + 0.3 x (P / Pmax) is calculated, where T is the driving time (seconds), Tmax is the maximum driving time of all candidate points, P is the point cleaning and transportation score, and Pmax is the highest cleaning and transportation score of all candidate points. The point with the highest comprehensive score S is selected as the next optimal work point. If the comprehensive scores are the same, the point with a higher cleaning and transportation score is preferred. Road congestion is also considered, and when the congestion index of a path exceeds 0.8, a 30% penalty coefficient is added to the driving time of the path in the calculation. The station coordinates (longitude λ2, latitude φ2) with the shortest driving time or the highest comprehensive score under the comprehensive conditions are finally determined as the next optimal work point. ) is specified as the first target point of the target collection and transportation vehicle. The preset Gaode map service interface is called through the RESTful API method to construct a batch path planning request. The request parameters include: the starting point coordinates (the longitude and latitude coordinates of the first target point), the terminal point coordinate set (the longitude and latitude coordinates of all remaining points in the coordinate data of the to-be-visited station), the driving strategy (set to the shortest time), and the vehicle type (set to a heavy truck according to the model of the target collection and transportation vehicle). An HTTP POST request is sent, carrying a JSON data body in a specified format, and the request timeout time is set to 3 seconds. The map service interface returns JSON format data containing all path schemes, and the expected driving time (unit: seconds), distance (unit: meters) and real-time traffic information of each path are extracted through a JSON parser. When the interface call fails, a backup offline path calculation module is started to estimate the driving time based on the Euclidean distance and the historical average driving speed (18 kilometers / hour).

[0122] The detailed information of the next optimal work point is added to the task list of the current vehicle. The task list is stored in a linked list structure, and each node contains five fields: site identification number, geographic coordinates, estimated arrival time, estimated work duration, and clean-up score. The estimated work duration is calculated based on the three-dimensional voxel data of the garbage in the garbage can, and the formula is t = 2 + V / 40, where t is the estimated work duration (minutes), V is the garbage volume (cubic meters), 2 is the basic operation time (minutes), and 40 is the standard clean-up rate (cubic meters / minute). The current position of the vehicle is updated to the coordinates of the newly added work point, and the cumulative travel time and cumulative travel distance are calculated. At the same time, the point information is removed from the to-be-visited site coordinate data structure, and the state field of the corresponding item is set to allocated using the mark deletion method. The total volume and total weight of the remaining to-be-visited sites are recalculated, and the remaining capacity information of the vehicle is updated. After the operation is completed, an operation log record is generated, including timestamp, vehicle number, work point information and remaining capacity data, which is stored in the log database to form the updated work path.

[0123] The loop iteration algorithm is executed, and steps S31 to S33 are packaged as a path planning function, which is repeatedly called to assign points. Before each iteration, two termination conditions are checked: the capacity limit of the target collection vehicle and the shift length limit. The capacity limit check formula is Vc + Vnext ≤ Vmax and Mc + Mnext ≤ Mmax, where Vc is the total volume of the currently allocated garbage, Vnext is the garbage volume of the next candidate point, Vmax is the maximum capacity of the vehicle, Mc is the total weight of the currently allocated garbage, Mnext is the garbage weight of the next candidate point, and Mmax is the maximum load of the vehicle. The shift length limit check formula is Tc + Tnext + Toperation ≤ Tmax, where Tc is the current cumulative time, Tnext is the driving time to the next point, Toperation is the estimated work duration of the next point, and Tmax is the maximum shift working time (480 minutes). When either condition is not met, the iteration process is terminated. The vehicle load state and remaining capacity are updated after each iteration, and the total travel distance and total time are recalculated. When the to-be-visited site coordinate data is empty or all points have been determined to be unable to be added, the iteration is also terminated. After all iterations are completed, the generated task list is marked as a feasible work sequence, and the total travel distance, total travel time and total work time of the sequence are calculated.

[0124] The average collection score of all points in the sequence is calculated, and then the key points with collection scores significantly higher than the average value are identified. For each key point, the difference between its position index i in the current sequence and the ideal position index i' is calculated, where the ideal position index is determined based on the collection score ranking. When Δi>3, the sequence rearrangement mechanism is triggered, and the key point is moved forward by at most 3 positions, while the affected points are moved backward. The rearrangement process adopts a local optimization strategy to ensure that the adjusted total travel distance increases by no more than 8% of the original plan. Considering the time window constraint, for garbage cans in commercial and public areas, the peak hours (7:00-9:00 and 17:00-19:00) avoidance rule is added, and the sensitive points in these time periods are adjusted to more suitable times. After all adjustments are completed, the indicators of the entire sequence are recalculated, the theoretically optimal collection sequence is generated, and the complete path navigation information is pushed to the on-board terminal device of the target collection vehicle through the wireless network.

[0125] Preferably, step S4 comprises the following steps:

[0126] Step S41: The theoretically optimal collection sequence is subjected to navigation path planning and segmented navigation instruction processing to obtain segmented navigation instructions;

[0127] Step S42: Based on each target site in the segmented navigation instructions, the structure characteristics are filled through intelligent analysis of the garbage three-dimensional structure voxel data to generate single-point fine operation notes;

[0128] Step S43: The segmented navigation instructions and the single-point fine operation notes of each site are combined and issued to the terminal of the target collection vehicle to manage the smart garbage can collection operation.

[0129] In the embodiment of the application, the site coordinate list in the theoretically optimal collection sequence is extracted as an ordered latitude and longitude array, which includes the current position coordinate of the collection vehicle as the starting point and all the to-be-visited site coordinates. The complete path is generated by calling the driving route planning interface of the map open platform. The interface call adopts a batch request method, with a maximum of 10 points processed each time, and more than 10 points processed in batches. The driving strategy parameter is set to the shortest time, the vehicle type parameter is set to a heavy truck, and the avoidance area parameter includes restricted areas and congested road sections. The JSON data returned by the interface includes the coordinate point set of the complete driving path, the turning prompt, the road name, the traffic direction, and other information. The navigation instruction segmentation processing is performed, and the continuous path is divided into multiple independent road segments according to the sites, and each road segment constitutes a segmented navigation instruction. The segmented navigation instruction includes: starting point coordinates, ending point coordinates, coordinate point set (one point every 10 meters), turning point set, estimated travel time, estimated travel distance, road segment road type, traffic state, and other key information.

[0130] A garbage 3D structure voxel data analysis is performed for each target site in the segmented navigation instruction. The latest garbage 3D structure voxel data of the target site is called from the database, and a 3D morphological feature extraction algorithm is performed. First, the height distribution histogram of the garbage filling is calculated to identify the filling height concentration interval; then, a density clustering analysis is performed to divide the voxels into high-density areas (p > 1.0 g / cm3), medium-density areas (0.5 ≤ p ≤ 1.0 g / cm3), and low-density areas (p < 0.5 g / cm3) according to the density values. Based on the clustering results, a filling characteristic label is generated, including heavy object concentration (bottom high-density proportion > 60%), light object majority (low-density area proportion > 70%), mixed filling (similar proportions of each density area), and foreign object presence (single-point density anomaly detection). The activity of the garbage is also analyzed according to the dynamic thermal decomposition degree data, and when the thermal decomposition degree data > 5, it is marked as high-temperature activity. Based on the above analysis, a structured single-point fine operation note is generated, including garbage volume estimate, weight estimate, physical property label, processing difficulty level (1-5 levels), and special processing prompt.

[0131] The segmented navigation instruction is matched with the single-point fine operation note. A unified operation instruction data structure is established, including a basic path information block, a site operation information block, and a control information block. For each target site, the segment data in the navigation instruction is associated with the fine operation note of the corresponding site to form a complete site operation instruction package. Through a 4G / 5G wireless communication network, the MQTT protocol is used to push the operation instruction package to the vehicle-mounted terminal device of the target collection and transportation vehicle. The push process sets the quality of service level to QoS2 (ensures delivery once and only once), and at the same time, 256-bit AES encryption is enabled to protect data security. After receiving the operation instruction package, the vehicle-mounted terminal automatically parses and loads the navigation information to the vehicle-mounted navigation, and displays the site fine operation note in a card layout on the driver's operation interface. The instruction execution state is monitored in real time, and when the vehicle arrives at the site, the terminal automatically displays the operation note of the corresponding site on top and reminds the driver to pay attention to special processing requirements through voice prompts. After the vehicle completes the site operation, the driver confirms the completion status on the terminal, automatically loads the navigation instruction and operation note of the next site, and realizes intelligent management throughout the whole process.

[0132] Especially important is that the filling structure features are identified and extracted based on the garbage 3D structure voxel data for each target site in the segmented navigation instruction, including

[0133] Based on the garbage 3D structure voxel data, a set of voxels with a density value less than a preset threshold of 0.3 kg / m3 and continuous in space is identified and extracted, which is marked as an internal structural cavity;

[0134] The ratio of the total volume of the internal structural cavity to the effective volume of the garbage can is calculated, and the vertical position distribution in the garbage three-dimensional structure voxel data is evaluated to assess the structural instability risk index due to gravity during vehicle driving bumps or dumping process;

[0135] The centroid position of the entire garbage pile is calculated using the density and three-dimensional coordinates of each voxel in the garbage three-dimensional structure voxel data, and it is judged whether the horizontal offset of the centroid and the geometric center of the garbage can is greater than 30% of the can radius. If so, it is determined that there is an unbalanced load, and an unbalanced state marker is obtained;

[0136] The impact force generated by the target collection and transportation vehicle when dumping the garbage can is quantified by combining the structural instability risk index and the unbalanced load imbalance state marker, and a predicted dynamic unloading impact value is generated;

[0137] According to the predicted dynamic unloading impact value, a single-point fine operation note is intelligently generated.

[0138] In the embodiment of the present application, the latest garbage three-dimensional structure voxel data of the target site is extracted from the database, including the three-dimensional coordinates (x, y, z) and the corresponding density value p of each voxel. A preset threshold p0=0.3 kg / m3 is set as the cavity judgment standard, which is obtained based on laboratory tests and represents the critical density value of loose garbage accumulation. A three-dimensional region growing algorithm is executed to scan layer by layer from the bottom of the garbage can, and when a voxel with a density value p< p0 is found, it is marked as a candidate cavity point. Subsequently, a 26-connectivity check is performed on each candidate cavity point (considering the face contact, edge contact and vertex contact of the voxel in three-dimensional space), and the voxels that are spatially continuous and adjacent and have a density value less than the threshold are classified into the same cavity set. Small cavities with less than 10 voxels are removed, and only large cavities with structural significance are retained. For each identified internal structural cavity, a unique identifier is assigned, and the voxel coordinate set, total volume, shape feature parameter and spatial position information in the garbage can are recorded. The total volume Vc of the internal structural cavity is calculated by multiplying the number of voxels marked as cavities by the volume of a single voxel v (cubic centimeters). The effective volume Vt (cubic meters) of the garbage can of this model is read from the device parameter library, and the cavity ratio r=Vc / Vt is calculated. At the same time, the vertical distribution characteristics of the cavity are analyzed, the garbage can height is equally divided into three layers (lower layer, middle layer and upper layer), and the cavity volume proportion of each layer rl, rm and ru is calculated. The structural instability risk assessment is performed, and the basic risk factor R0= r x 10 is first calculated, with a value range of 0 to 10. Then according to the vertical distribution correction, when the lower layer cavity proportion rl>0.5, the risk coefficient increases by 30%; when the middle layer cavity proportion rm>0.4, the risk coefficient increases by 50%; when there is a continuous cavity penetrating more than two layers, the risk coefficient increases by 70%. The final structural instability risk index Rs is calculated by the formula Rs=R0×(1+0.3×I1+0.5×I2+0.7×I3), where I1, I2 and I3 are Boolean indicator variables, taking the value 1 when the corresponding conditions are met, and 0 otherwise.

[0139] The centroid position of the entire garbage pile is calculated based on the three-dimensional structure voxel data of the garbage. For each non-cavity voxel i, its three-dimensional coordinates (xi, yi, zi) and density value pi are obtained. The centroid coordinate calculation formula is: xc=∑(xi×pi×v) / M, yc=∑(yi×pi×v) / M, zc=∑(zi×pi×v) / M, where v is the volume of a single voxel, and M=∑(pi×v) is the total mass of the garbage. The calculated centroid coordinates (xc, yc, zc) take the center of the bottom of the garbage can as the origin, the right direction as the positive direction of the x-axis, the forward direction as the positive direction of the y-axis, and the upward direction as the positive direction of the z-axis. The horizontal offset of the centroid is calculated as where (x0, y0) is the geometric center coordinate of the trash can. The radius R of the trash can is obtained from the device parameter library, and the relative offset ratio p = d / R is calculated. When p > 0.3 (i.e., the horizontal offset is greater than 30% of the radius of the can), it is determined that the trash can is off-loaded, and the off-load imbalance state flag B = 1 is set; otherwise, B = 0 is set. The off-load direction angle θ = arctan((yc-y0) / (xc-x0)) is also recorded.

[0140] The static base impact force F0 = M x g x h / t is calculated, where M is the total mass of the garbage (kg), g is the acceleration of gravity (9.8 m / s2), h is the dumping height of the trash can (m), and t is the typical dumping time (s). When there is an internal structural cavity, the additional impact caused by the collapse effect is considered, and the structural instability gain coefficient Ks = 1 + 0.15 x Rs is calculated. When there is an off-load, the lateral moment caused by the unbalanced rotation is considered, and the off-load gain coefficient Kb = 1 + 0.2 x B x p is calculated, where p is the relative offset ratio. Considering both factors, the dynamic unloading impact value F = F0 x Ks x Kb is estimated. The F value is mapped to a 5-level quantization scale: F < 500 Newton is level 1 (light impact), 500 ≤ F < 1000 Newton is level 2 (moderate impact), 1000 ≤ F < 2000 Newton is level 3 (stronger impact), 2000 ≤ F < 3500 Newton is level 4 (intense impact), and F ≥ 3500 Newton is level 5 (extremely strong impact).

[0141] For sites with an impact value of level 1, a standard operation prompt is generated for regular dumping; for level 2 sites, a slow dumping prompt is generated and marked with a yellow warning; for level 3 sites, a segmented dumping prompt is generated, suggesting a 45-degree inclination for 3 seconds, and then continuing to dump after the contents are stable; for level 4 sites, a small-angle multiple dumping prompt is generated, suggesting repeated small-angle shaking before complete dumping; for level 5 sites, a special processing prompt is generated, marked with a red warning, and it is recommended to use an auxiliary push rod to assist in unloading. When the off-load imbalance state flag B = 1, an off-load direction prompt is added to the operation note, such as right side gravity offset, adjust the position to the left side before dumping. When the internal cavity is located in the middle and lower layers and the cavity ratio r > 0.2, a special reminder is added to pay attention to the sudden sliding of the material during the dumping process. These fine operation notes are integrated into structured text and diagrams for the collection and transportation personnel to refer to and execute.

[0142] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, the scope of the present application being defined by the appended claims rather than the above description, and it is intended to encompass all variations falling within the meaning and scope of the equivalent elements of the application file.

[0143] The foregoing is considered as illustrative only of the principles of the application. Numerous modifications and changes will readily occur to those skilled in the art, and it is intended to embrace all such modifications and changes that fall within the scope of the application. Accordingly, the application is not to be restricted in scope to the specific embodiments disclosed herein but is to be accorded the full scope that the principles and novel features request appropriately granted.

Claims

1. An implementation method of a smart city Internet of Things device data collection management platform, characterized in that, The method comprises the following steps: Step S1: deploying Internet of Things sensing devices inside the garbage can, and constructing an Internet of Things front-end sensing network; Collecting data of incremental throwing events in the garbage can through the Internet of Things front-end sensing network, and analyzing the structural voxel changes caused by the incremental throwing events to generate garbage three-dimensional structural voxel data; Step S2: calculating the total filling rate of garbage according to the garbage three-dimensional structural voxel data; evaluating dynamic thermal decomposition degree data according to the garbage three-dimensional structural voxel data, and calculating the in-can state urgency according to the total filling rate of garbage to obtain a dynamic urgency index; According to the dynamic urgency index, the supervision weighted collection and transportation urgency is processed, and then the to-be-processed site cluster is identified; Step S3: extracting the to-be-visited site coordinate data in the to-be-processed site cluster, and dynamically planning the theoretical optimal collection and transportation sequence of the target collection and transportation vehicle; Step S4: processing the theoretical optimal collection and transportation sequence by segmented navigation instruction, and adding single-point fine work notes through the garbage three-dimensional structural voxel data to manage the intelligent garbage can collection and transportation work.

2. The implementation method of the smart city IoT device data collection management platform according to claim 1, characterized in that, Step S1 comprises the following steps: Step S11: deploying Internet of Things sensing devices inside the garbage can in the smart city, and jointly accessing an edge computing unit with at least four general input and output interfaces and one analog-digital conversion interface, thereby constructing an Internet of Things front-end sensing network; wherein the Internet of Things sensing devices comprise an ultrasonic sensor array, a pressure sensor, an infrared thermal imaging sensor, a gas sensor, and a temperature and humidity sensor; Step S12: reading the voltage signal of the pressure sensor in the Internet of Things front-end sensing network in real time, and obtaining garbage real-time filling total weight data after analog-digital conversion; Step S13: based on the garbage real-time filling total weight data, marking an incremental throwing event at each time when the weight changes by more than 100 grams; Step S14: based on the time of the incremental throwing event, using the ultrasonic sensor array in the Internet of Things front-end sensing network to obtain the echo time of each sensor to the garbage surface to obtain the original echo time; Step S15: converting the original echo time into the spatial rectangular coordinates of the corresponding sensor according to the preset sound speed value, thereby determining the surface height coordinate point inside the garbage can; Step S16: analyzing the structural voxel changes of the throwing event according to the surface height coordinate point and the garbage real-time filling total weight data to generate garbage three-dimensional structural voxel data.

3. The implementation method of the smart city IoT device data collection management platform according to claim 2, characterized in that, The analysis of the structural voxel changes of the throwing event according to the surface height coordinate point and the garbage real-time filling total weight data in step S16 comprises: Based on the incremental throwing event, calculating the ratio of the volume increment to the weight increment of each throwing event through the surface height coordinate point and the garbage real-time filling total weight data, evaluating the density of different batches of garbage, and obtaining layered garbage density values; Obtaining historical surface height time series data; Using the historical surface height time series data and the preloaded garbage can three-dimensional geometric parameters in the edge computing unit to perform coordinate mapping and basic voxel grid calibration in real time, and using the layered garbage density values to weight the density of the corresponding depth voxels to generate a density layered voxel model; The lower layer voxel grid is divided for the density layered voxel model, and then the gravity sedimentation effect is simulated and corrected, that is, the density value of the layered garbage corresponding to the lower layer voxel grid is compressed by 1-5% to obtain the garbage three-dimensional structure voxel data.

4. The implementation method of the smart city IoT device data collection management platform according to claim 1, characterized in that, In step S2, the dynamic thermal decomposition degree data is evaluated according to the garbage three-dimensional structure voxel data, and the barrel state urgency is calculated according to the total filling rate of the garbage, including: The temperature distribution information inside the garbage can is collected by the infrared thermal imaging sensor in the Internet of Things front-end sensing network to obtain the barrel infrared temperature data. The garbage three-dimensional structure voxel data is used as a spatial reference, and pixel-level registration and superposition are performed according to the barrel infrared temperature data to generate three-dimensional temperature tag voxel data. The ambient temperature and ambient humidity are obtained by the temperature and humidity sensor in the Internet of Things front-end sensing network. The dynamic thermal decomposition degree data is evaluated according to the three-dimensional temperature tag voxel data and the ambient temperature. The environmental risk compensation is performed on the dynamic thermal decomposition degree data based on the ambient humidity, and the barrel state urgency is calculated according to the total filling rate of the garbage, to obtain the dynamic urgency index.

5. The implementation method of the smart city IoT device data collection management platform according to claim 4, characterized in that, The dynamic thermal decomposition degree data is evaluated according to the three-dimensional temperature tag voxel data and the ambient temperature, including: The continuous area with a temperature higher than the ambient temperature by more than 5°C is extracted from the three-dimensional temperature tag voxel data as an effective heating area. The average filling density of the garbage in the effective heating area is calculated by the garbage three-dimensional structure voxel data. When the average filling density in the effective heating area is greater than a preset first content threshold, the content risk weight factor of the effective heating area in the garbage can is set to 1.5, which is recorded as a high-risk organic matter weight. When the average filling density in the effective heating area is less than a preset second content threshold, the content risk weight factor of the effective heating area in the garbage can is set to 0.5, which is recorded as a low-risk organic matter weight; wherein the preset first content threshold is greater than the preset second content threshold. The voxel activity score processing is performed on the effective heating area according to the high-risk organic matter weight and the low-risk organic matter weight, and then the highest activity score in the voxel is extracted as the dynamic thermal decomposition degree data of the garbage decomposition activity degree.

6. The implementation method of the smart city IoT device data collection management platform according to claim 4, characterized in that, The environmental risk compensation is performed on the dynamic thermal decomposition degree data based on the ambient humidity, and the barrel state urgency is calculated according to the total filling rate of the garbage, to obtain the dynamic urgency index, including: The gain compensation is performed on the dynamic thermal decomposition degree data according to the ambient humidity to generate compensated thermal decomposition degree data. The real-time gas concentration value is obtained by the gas sensor in the Internet of Things front-end sensing network. The local weather data is obtained, and the current environmental wind speed is extracted as the local environmental wind speed value. It is judged whether the real-time gas concentration value is greater than 50ppm and the local environmental wind speed value is less than 1m / s, if yes, the influence coefficient is set to 1.4, if not, it is set to 1.0, to obtain the odor diffusion influence coefficient. The public health sensitivity processing is performed on the compensated thermal decomposition degree data by the odor diffusion influence coefficient, and the barrel state urgency is calculated according to the total filling rate of the garbage, to obtain the dynamic urgency index.

7. The implementation method of the smart city IoT device data collection management platform according to claim 1, characterized in that, In step S3, the supervision weighted clean-up urgency processing is performed according to the dynamic urgency index, including: Obtaining a functional area category label of the garbage can; and matching a supervision weight coefficient according to the functional area category label; wherein the functional area category label comprises a commercial area, a residential area, and a public place; Weighted clean-up urgency calculation is performed according to the supervision weight coefficient and the dynamic urgency index, and priority score normalization processing is performed to generate clean-up score data; A collection and transportation starting threshold is set to 75, garbage cans with clean-up score data greater than 75 are screened to obtain a high-priority site list, otherwise a sub-priority site list is obtained; The site with the highest clean-up score in the high-priority site list is selected as the initial center of regional clustering, serving as a cluster generation center point; A range with a radius of 1 km is drawn on the map with the cluster generation center point as the center, and all high-priority sites and part of the sub-priority sites falling within the range are collected to obtain a potential cluster member set; The estimated total volume and total weight of the garbage are calculated according to the three-dimensional structure voxel data of the garbage to obtain the estimated load of the cluster; The target collection and transportation vehicle closest to the cluster generation center point is matched; The estimated load of the cluster is compared with the loadable weight of the target collection and transportation vehicle. If it exceeds, the site with the lowest urgency is removed from the potential cluster member set until the condition is met to generate a site cluster to be processed.

8. The implementation method of the smart city IoT device data collection management platform according to claim 7, characterized in that, Step S3 includes the following steps: Step S31: The garbage can with the highest priority in the to-be-visited site coordinate data is taken as the first target point of the target collection and transportation vehicle, and then the real-time driving time of all remaining points in the to-be-visited site coordinate data is queried through a preset map service interface to obtain candidate point time consumption; Step S32: The point with the shortest time consumption is selected from the candidate point time consumption to determine the next optimal work point of the vehicle; Step S33: The next optimal work point is added to the task list of the current vehicle, and the point is removed from the to-be-visited site coordinate data to obtain an updated work path; Step S34: Steps S31 to S33 are repeated until the capacity or shift time limit of the target collection and transportation vehicle is met, and the updated work path is taken as a feasible work sequence; Step S35: The feasible work sequence is subjected to secondary weight adjustment of priority through the clean-up score data to obtain a theoretically optimal collection and transportation sequence.

9. The implementation method of the smart city IoT device data collection management platform according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: The theoretically optimal collection and transportation sequence is subjected to navigation path planning, and segmented navigation instruction processing is performed to obtain segmented navigation instructions; Step S42: Based on each target site in the segmented navigation instructions, structure features are filled through intelligent analysis of the three-dimensional structure voxel data of the garbage to generate single-point fine work annotations; Step S43: The segmented navigation instructions and the single-point fine work annotations of each site are combined and sent to the terminal of the target collection and transportation vehicle to manage the intelligent garbage can collection and transportation work.

10. A smart city IoT device data collection management platform, characterized in that, It includes: The data layer, the service layer and the application layer, the data layer output end sets the service layer, the service layer output end sets the application layer, the data layer is used for collecting wisdom city information and carries out analysis, the service layer is used for providing different services to meet the access demand of the application layer, the application layer realizes the access of intelligent terminal equipment through the access interface, the data layer is used for executing the implementation method of the wisdom city internet of things equipment data collection management platform as claimed in claim 1, the data layer in the wisdom city internet of things equipment data collection management platform includes: The internet of things data collection module is used for deploying internet of things sensing devices inside the garbage can and constructing an internet of things front-end sensing network;Through the internet of things front-end sensing network, data of incremental throwing events in the garbage can is collected, and the change of structural voxels in the garbage can caused by the incremental throwing events is analyzed to generate garbage three-dimensional structural voxel data; The dynamic risk assessment module is used for calculating the total filling rate of garbage according to the garbage three-dimensional structural voxel data;According to the garbage three-dimensional structural voxel data, the dynamic thermal decomposition degree data is evaluated, and the in-can state emergency degree is calculated according to the total filling rate of garbage, to obtain the dynamic urgency index;According to the dynamic urgency index, the supervision weighted collection urgency processing is carried out, and then the to-be-processed site cluster is identified; The collection path planning module is used for extracting the to-be-visited site coordinate data in the to-be-processed site cluster, and dynamically planning the theoretical optimal collection sequence of the target collection vehicle; The intelligent job management module is used for segmenting the theoretical optimal collection sequence for navigation instruction processing, and performing single-point fine job annotation through the garbage three-dimensional structural voxel data, to realize the intelligent management of garbage can collection in wisdom city.

Images