Intelligent dustbin full load prediction and notification method and system and storage medium

Abstract

The invention relates to an intelligent dustbin full load prediction and notification method and system and a storage medium. The method comprises the steps that weight and volume changes of garbage in a dustbin are obtained; performing preprocessing operation on the acquired data; according to the obtained historical and real-time data, the predicted full load time of the dustbin is calculated through a linear regression prediction model, and whether a notification mechanism needs to be triggered or not is judged; and when it is determined that the garbage can is about to be fully loaded according to the predicted full-load time, a notification mechanism is automatically triggered, and the content of a notification comprises the specific position of the garbage can, the predicted full-load time and a cleaning suggestion. According to the invention, the functions of real-time monitoring, accurate prediction and intelligent notification cleaning are realized.

Description

Technical Field

[0001] The present invention relates to the fields of Internet of Things technology, artificial intelligence algorithm and garbage disposal technology, and in particular to a method, system and storage medium for predicting and notifying full load of an intelligent garbage bin. Background Art

[0002] With the rapid advancement of global urbanization, the urban population continues to expand, and with it comes a sharp increase in the amount of urban domestic waste. This trend poses a severe challenge to the traditional urban waste disposal system, especially in the collection, transportation and treatment of waste. The traditional waste disposal method mainly relies on manual inspections and regular cleaning. Although it maintains the basic cleanliness of the city to a certain extent, its inherent limitations are becoming increasingly prominent and cannot meet the urgent needs of modern cities for efficient, intelligent and environmentally friendly waste disposal systems.

[0003] Specifically, traditional garbage disposal methods have the following main disadvantages:

[0004] 1. Limitations of manual inspection mode

[0005] Inefficiency and high cost: Manual inspections require a lot of human resources and time costs. Due to manpower constraints, the inspection frequency and coverage are limited, making it difficult to detect and respond to full trash bins in a timely manner, resulting in frequent garbage accumulation.

[0006] Delayed response: Manual inspections often rely on periodic checks and cannot achieve real-time monitoring of the status of garbage bins, making it difficult to deal with full-load problems in the first place, affecting urban environmental sanitation and residents’ quality of life.

[0007] 2. Lack of scientific forecasting and scheduling mechanisms

[0008] Unreasonable resource allocation: The traditional scheduled cleaning model lacks accurate predictions of garbage generation and garbage bin fullness, resulting in the allocation of cleaning resources often being based on experience, which can easily lead to excessive concentration or waste of resources.

[0009] Lack of planning: The lack of a scientific forecasting mechanism makes garbage collection work lack foresight and planning, making it difficult to flexibly adjust the frequency and route of cleaning according to actual needs, affecting the efficiency and effectiveness of garbage disposal.

[0010] 3. Environmental pollution and public health risks

[0011] Land occupation and ecological damage: Long-term accumulation of garbage not only occupies precious land resources, but may also cause soil pollution and ecological damage due to improper disposal.

[0012] Secondary pollution: Pollutants such as leachate and foul-smelling gases generated during garbage accumulation pose a serious threat to groundwater, air quality and human health, and aggravate urban environmental pressures and public health risks.

[0013] In view of the various drawbacks of the above-mentioned traditional garbage disposal methods, it is particularly important to develop an intelligent garbage bin full load prediction and notification method, system and storage medium that can realize real-time monitoring, accurate prediction and intelligent scheduling of the garbage bin status. Summary of the invention

[0014] The purpose of the present invention is to provide a smart garbage bin full load prediction and notification method, system and storage medium, which can realize the functions of real-time monitoring, accurate prediction and intelligent notification and cleaning.

[0015] In a first aspect, a method for predicting and notifying a full load of an intelligent garbage bin according to the present invention comprises the following steps:

[0016] Obtain the weight and volume changes of garbage in the garbage bin;

[0017] Perform preprocessing operations on the acquired data;

[0018] Based on the historical and real-time data obtained, a linear regression prediction model is used to calculate the expected full load time of the trash bin and determine whether a notification mechanism needs to be triggered; wherein the linear regression prediction model is:

[0019] T=β 0 +β 1 W+β 2 ·V+∈

[0020] Where T is the estimated full load time, W is the weight of the garbage in the trash bin, V is the volume of the garbage in the trash bin, and β 0 , β 1 , β 2 They are the parameters of the linear regression prediction model, which are learned through training data, and ∈ is the random error term;

[0021] When it is determined that the trash bin is about to be full based on the estimated full load time, a notification mechanism is automatically triggered, where the content of the notification includes the specific location of the trash bin, the estimated full load time and cleaning suggestions.

[0022] Optionally, it also includes:

[0023] Obtain the concentration of harmful gases in the garbage bin;

[0024] Obtain image information of garbage in the garbage bin, and identify the type of garbage based on the image information;

[0025] Determine the hazard level based on the concentration of harmful gases and the type of garbage;

[0026] Calculate a comprehensive score based on the hazard level and estimated full load time;

[0027] When the comprehensive score is greater than the preset score threshold, the notification mechanism is automatically triggered. When the bin is not fully loaded based on the expected full load time, the harmful gas concentration and garbage type identification functions are further introduced to more comprehensively evaluate the hazard level of the bin. Through the comprehensive scoring mechanism, it is determined whether to trigger the notification mechanism, so that bins with high hazard levels can be processed first, effectively reducing environmental pollution and public health risks.

[0028] Optionally, the specific method of determining the hazard level is:

[0029] If the concentration of the harmful gas is greater than or equal to the first preset concentration threshold, the hazard level is high;

[0030] If the concentration of the harmful gas is less than the first preset concentration threshold and greater than or equal to the second preset concentration threshold, and the type of the garbage is the preset hazardous type, the hazard level is high, otherwise the hazard level is medium, wherein the first preset concentration threshold is greater than the second preset concentration threshold;

[0031] If the concentration of harmful gases is less than the second preset concentration threshold, the hazard level is low. The specific criteria for judging the hazard level are clarified, including the concentration threshold of harmful gases and the preset hazard types of garbage. This standard makes the judgment of the hazard level more objective and accurate, and provides a reliable basis for the comprehensive scoring and determination of cleaning priorities.

[0032] Optionally, the calculation formula of the comprehensive score is as follows:

[0033] Comprehensive score = α*(1 / estimated full load time)+β*value corresponding to the hazard level);

[0034] Among them, α represents the weight of the predicted full load time, and β represents the weight of the hazard level. By comprehensively considering the predicted full load time and the hazard level, a calculation formula for the comprehensive score was developed. This formula can fully reflect the urgency and hazard level of the garbage bin, providing a scientific basis for the reasonable arrangement of the cleaning work.

[0035] Optionally, the harmful gas includes at least one of ammonia, hydrogen sulfide, sulfur dioxide, nitrogen-containing compounds and volatile organic compounds. It is clarified that harmful gases include ammonia, hydrogen sulfide, sulfur dioxide, nitrogen-containing compounds and volatile organic compounds, etc. These gases are common pollutants in the garbage accumulation process. By monitoring the concentration of these gases, it is possible to timely discover possible environmental pollution problems in the garbage bin, providing a basis for taking corresponding treatment measures.

[0036] Optionally, preprocess the acquired data, specifically:

[0037] Clean the data to remove invalid data and outliers;

[0038] De-noising the data;

[0039] Compress the data. By cleaning, denoising and compressing the acquired data, the accuracy and availability of the data are improved. This step is the basis for subsequent prediction and evaluation work and is of great significance for ensuring the stability and accuracy of the entire system.

[0040] Optionally, based on the acquired historical and real-time data, a linear regression prediction model is used to calculate the estimated full-load time of the trash bin, specifically:

[0041] Build a linear regression prediction model that learns from historical data and understands the relationship between the weight and volume changes of garbage in the garbage bin and the full load time;

[0042] After the linear regression prediction model is built, the linear regression prediction model is trained using historical data to solve the parameter β of the linear regression prediction model. 0 , β 1 , β 2 , so that the prediction error is minimized;

[0043] Receive real-time data and use the real-time data to update and correct the linear regression prediction model;

[0044] After the linear regression prediction model is updated, the estimated full-load time of the trash bin is calculated using the linear regression prediction model. The construction, training, and update process of the linear regression prediction model is described in detail. Through this process, accurate prediction of the full-load time of the trash bin can be achieved, providing reliable data support for the intelligent notification mechanism.

[0045] Optionally, notification is made through at least one of SMS, APP push, and email. Multiple notification methods such as SMS, APP push, and email are provided to meet the needs of different user groups. This flexibility enables the notification mechanism to more effectively cover relevant personnel and improve the efficiency of information delivery and response speed.

[0046] In a second aspect, a smart trash bin full load prediction and notification system according to the present invention comprises a memory and a controller, wherein the memory stores a computer-readable program, and when the computer-readable program is called by the controller, it can execute the steps of the smart trash bin full load prediction and notification method according to the present invention.

[0047] In a third aspect, a storage medium described in the present invention stores a computer-readable program therein, and when the computer-readable program is called by a controller, it can execute the steps of the intelligent garbage bin full load prediction and notification method described in the present invention.

[0048] Beneficial effects of the present invention:

[0049] (1) Real-time monitoring and accurate prediction:

[0050] By integrating sensors and other technical means, real-time monitoring of garbage weight, volume and harmful gas concentration in the trash bin is achieved, improving the accuracy and timeliness of data acquisition.

[0051] Using algorithms such as linear regression prediction models, the expected full load time of the garbage bin can be accurately predicted based on historical and real-time data, providing a scientific basis for garbage collection work.

[0052] (2) Intelligent notification and efficient scheduling:

[0053] When the trash bin is expected to be full, the notification mechanism is automatically triggered to promptly notify relevant personnel through SMS, APP push, email and other methods, thus achieving rapid transmission of information.

[0054] The notification content includes the specific location of the trash bin, the estimated full load time and cleaning suggestions, which helps relevant personnel respond quickly, arrange cleaning work reasonably, and improve the efficiency of garbage disposal.

[0055] (3) Hazard level assessment and priority determination:

[0056] By analyzing the concentration of harmful gases and the type of garbage in the trash bin, the hazard level is determined, and a comprehensive score is calculated based on the hazard level and the expected full load time, providing a basis for determining the cleaning priority.

[0057] This will help prioritize the disposal of high-hazard level trash bins, reduce environmental pollution and public health risks, and improve urban environmental sanitation.

[0058] (4) Resource optimization and cost saving:

[0059] The intelligent trash bin full load prediction and notification method can realize the rational allocation and efficient utilization of garbage cleaning resources, avoiding excessive concentration or waste of resources.

[0060] By reducing the frequency of unnecessary inspections and cleanings, the investment in manpower, material and financial resources is reduced, saving costs.

[0061] (5) Environmental friendliness and sustainable development:

[0062] The application of intelligent trash bin full load prediction and notification methods can help reduce problems such as land occupation, ecological damage and secondary pollution caused by long-term garbage accumulation.

[0063] By improving the efficiency and effectiveness of garbage disposal, the improvement and sustainable development of the urban environment are promoted.

[0064] In summary, the intelligent garbage bin full load prediction and notification method, system and storage medium proposed in the present invention have demonstrated significant technical effects in real-time monitoring, accurate prediction, intelligent notification, hazard level assessment, resource optimization, cost savings and environmental friendliness, providing strong technical support for the upgrading and transformation of modern urban garbage disposal systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 is a flow chart of the intelligent trash bin full load prediction and notification method described in Embodiment 1 of the present application;

[0066] Figure 2 is a flow chart of the intelligent trash bin full load prediction and notification method described in Embodiment 2 of the present application;

[0067] Figure 3 It is a flow chart of constructing a linear regression model in an embodiment of the present application;

[0068] Figure 4 It is a principle block diagram of the intelligent trash bin full load prediction and notification system described in the embodiment of the present application;

[0069] Figure 5 It is a functional module diagram of the intelligent garbage bin full load prediction and notification system described in Example 1 of the present application;

[0070] Figure 6 It is a functional module diagram of the intelligent trash bin full load prediction and notification system described in Example 2 of the present application. DETAILED DESCRIPTION

[0071] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, not for limiting the scope of protection of the present invention.

[0072] Embodiment 1

[0073] like Figure 1 As shown, in an embodiment of the present application, a method for predicting and notifying a full load of an intelligent garbage bin includes the following steps:

[0074] Obtain the weight and volume changes of garbage in the garbage bin;

[0075] Perform preprocessing operations on the acquired data;

[0076] Based on the historical and real-time data obtained, a linear regression prediction model is used to calculate the expected full load time of the trash bin and determine whether the notification mechanism needs to be triggered; the linear regression prediction model is:

[0077] T=β 0 +β 1 W+β 2 ·V+∈

[0078] Where T is the estimated full load time, W is the weight of the garbage in the trash bin, V is the volume of the garbage in the trash bin, and β 0 , β 1 , β 2 They are the parameters of the linear regression prediction model, which are learned through training data, and ∈ is the random error term;

[0079] When it is determined that the trash can is about to be full based on the estimated full load time, a notification mechanism is automatically triggered, where the content of the notification includes the specific location of the trash can, the estimated full load time and cleaning suggestions.

[0080] like Figure 4 As shown, in a possible embodiment, a smart garbage bin full load prediction and notification system includes a memory and a controller, wherein the memory stores a computer readable program, and when the computer readable program is called by the controller, it can execute the steps of the smart garbage bin full load prediction and notification method as described in the present invention.

[0081] like Figure 5 As shown, in a possible embodiment, a smart garbage bin full load prediction and notification system is divided according to functional modules, including an Internet of Things sensor module, a data acquisition and preprocessing module, an AI prediction algorithm module and a notification module. Among them:

[0082] IoT sensor module: The IoT sensor module contains high-precision weight sensors and volume sensors, which are installed at different positions inside the trash bin. They can monitor the weight and volume changes of garbage in the trash bin in real time, thereby more accurately reflecting the filling level of the trash bin.

[0083] Data acquisition and preprocessing module: responsible for receiving data from the IoT sensor module and performing preprocessing operations such as data cleaning, denoising, and compression to ensure the accuracy and reliability of the data.

[0084] AI prediction algorithm module: It uses advanced linear regression prediction algorithm, combined with historical data and real-time data, to accurately predict the filling status of the garbage bin. At the same time, the AI ​​prediction algorithm module also has a self-learning function, which can continuously optimize the linear regression prediction model and improve the prediction accuracy.

[0085] Notification module: When the AI ​​prediction algorithm module predicts that the trash bin is about to be full, the notification module will automatically trigger the notification mechanism and notify the cleaners through SMS, APP push, email, etc. The notification content includes the specific location of the trash bin, the estimated full load time, and cleaning suggestions.

[0086] In summary, an intelligent garbage bin full load prediction and notification system, through the close cooperation of multiple links such as the IoT sensor module, data acquisition and preprocessing module, AI prediction algorithm module and notification module, realizes real-time monitoring of the weight and volume changes of garbage in the garbage bin, and determines whether to trigger the notification mechanism, and can promptly notify the cleaning staff to carry out garbage removal work. This system not only improves the efficiency and quality of garbage disposal, but also reduces the workload of cleaning staff, making a positive contribution to the beautification and protection of the urban environment.

[0087] The following is a detailed description of a method for predicting and notifying a full load of an intelligent garbage bin in an embodiment of the present application. The method comprises the following steps:

[0088] Step 1: The IoT sensor module includes a weight sensor and a volume sensor, which monitors the weight and volume changes of garbage in the trash bin in real time and sends the monitored data to the data acquisition and preprocessing module.

[0089] In step one, the IoT sensor module is installed in the trash bin, which can accurately sense the weight and volume changes of the garbage inside the trash bin. The IoT sensor module collects various data in the trash bin in real time through built-in sensitive components. For example, the weight sensor can measure the total weight of the garbage, and the volume sensor can estimate the size of the space occupied by the garbage. These data are crucial for subsequent analysis and prediction.

[0090] Once the sensors collect data, they will immediately transmit the data to the data acquisition and preprocessing module through wireless communication technologies such as Wi-Fi, LoRa or NB-IoT. As the starting point of data processing, the data acquisition and preprocessing module is responsible for receiving the raw data from each sensor and performing preliminary sorting and summarization.

[0091] Step 2: The data acquisition and preprocessing module performs preprocessing operations such as cleaning, denoising, and compression on the data, and transmits the processed data to the AI ​​prediction algorithm module.

[0092] The data acquisition and preprocessing module is a key link in the data processing process. Since sensor data may be affected by many factors, such as environmental factors, equipment failures, etc., there may be problems such as noise and outliers in the data. Therefore, the data acquisition and preprocessing module needs to perform a series of preprocessing operations on the received data to ensure the accuracy and reliability of the data.

[0093] First, the data acquisition and preprocessing module cleans the data to remove invalid data and outliers. This usually includes checking the integrity, rationality, and consistency of the data. For example, if the data from a sensor suddenly becomes abnormally large or small, then the data is likely invalid and needs to be removed.

[0094] Secondly, the data acquisition and preprocessing module will perform denoising on the data. Since the sensor data may be affected by various interference factors during the transmission process, there will be noise in the data. The purpose of denoising is to eliminate these noises and improve the signal-to-noise ratio of the data.

[0095] Finally, the data acquisition and preprocessing module will also compress the data. Since sensor data usually contains a lot of redundant information, compression can reduce the storage space and transmission bandwidth of the data. At the same time, compression also helps to improve data processing efficiency.

[0096] The preprocessed data will be transmitted to the AI ​​prediction algorithm module to provide reliable data support for subsequent prediction and analysis work.

[0097] Step 3: The AI ​​prediction algorithm module uses the linear regression prediction algorithm to calculate the estimated full load time of the trash bin based on historical and real-time data, and determines whether the notification mechanism needs to be triggered. If necessary, the notification content is generated and sent to the notification module.

[0098] The AI ​​prediction algorithm module is the core part of the whole system. The AI ​​prediction algorithm module uses machine learning technology to conduct in-depth analysis of historical and real-time data to predict the expected full load time of the garbage bin. To achieve this goal, the AI ​​prediction algorithm module first builds a linear regression prediction model. By learning from historical data, the linear regression prediction model can grasp the relationship between the weight and volume changes of garbage in the garbage bin and the full load time.

[0099] The AI ​​prediction algorithm module then uses real-time data to update and calibrate the linear regression prediction model to improve the accuracy of the prediction. After the AI ​​prediction algorithm module calculates the estimated full load time of the trash bin, it compares it with the preset time threshold. If the estimated full load time is less than or equal to the preset time threshold, the AI ​​prediction algorithm module will determine that the notification mechanism needs to be triggered and generate the corresponding notification content.

[0100] The notification content usually includes the location information of the trash bin, the estimated full load time, and the contact information of the cleaners, etc. This information is crucial for cleaners, helping them to understand the status of the trash bin in a timely manner and take appropriate measures.

[0101] Finally, the AI ​​prediction algorithm module will send the generated notification content to the notification module so that subsequent notification work can proceed smoothly.

[0102] Step 4: When the AI ​​prediction algorithm module predicts that the trash bin is about to be full, the notification module will automatically trigger the notification mechanism and notify the cleaners through SMS, APP push, email, etc. The notification content includes the specific location of the trash bin, the expected full load time, and cleaning suggestions.

[0103] like Figure 3 As shown, in a possible embodiment, the specific method of constructing the linear regression prediction model is as follows:

[0104] (31) Construction of linear regression prediction model:

[0105] The AI ​​prediction algorithm module first builds a linear regression prediction model. The core of this linear regression prediction model is to learn the linear relationship between the weight (W) and volume (V) of the garbage in the garbage bin and the full load time (T) through historical data. The linear regression prediction model formula is as follows:

[0106] T=β 0 +β 1 W+β 2 ·V+∈

[0107] Where T is the estimated full load time, W is the weight of the garbage in the trash bin, V is the volume of the garbage in the trash bin, and β 0 , β 1 , β 2 They are the parameters of the linear regression prediction model, which are learned through training data, and ∈ is the random error term;

[0108] (32) Model training and real-time updating:

[0109] After the linear regression prediction model is built, the AI ​​prediction algorithm module will use historical data to train the linear regression prediction model and solve the parameter β of the linear regression prediction model through the least squares method or other optimization algorithms. 0 , β 1 , β 2 , so that the prediction error is minimized.

[0110] Subsequently, real-time data is continuously received and used to update and calibrate the linear regression prediction model. This process usually includes steps such as preprocessing of new data, feature extraction, and online adjustment of model parameters to ensure that the linear regression prediction model can reflect the latest garbage filling trends, thereby improving the accuracy and timeliness of the prediction.

[0111] (33) Prediction and judgment:

[0112] When the linear regression prediction model is updated, the AI ​​prediction algorithm module will use the linear regression prediction model to predict the estimated full load time of the current trash bin. After the prediction is completed, the AI ​​prediction algorithm module will compare the predicted full load time T with the preset time threshold T0. If the predicted full load time T is less than or equal to the preset time threshold T0, it is determined that the notification mechanism needs to be triggered.

[0113] (34) Notification content generation and sending:

[0114] Once it is determined that the notification mechanism needs to be triggered, the AI ​​prediction algorithm module will generate notification content based on the preset notification template. The notification content usually includes key information such as the location information of the trash can (such as latitude and longitude coordinates or specific address), the expected full load time T, and the contact information of the cleaner. Subsequently, the AI ​​prediction algorithm module will send the generated notification content to the notification module so that subsequent notification work can proceed smoothly.

[0115] Through the above steps, the AI ​​prediction algorithm module not only realizes the accurate prediction of the expected full load time of the garbage bin, but also intelligently triggers the notification mechanism, providing strong technical support for the city's environmental sanitation management.

[0116] Specific examples are as follows:

[0117] Assume that there is the following historical data: the garbage weight W and volume V of a certain trash bin at different time points in the past week, as well as the corresponding full load time T. For example, at 8 a.m. one day, the garbage weight of the trash bin is 100 kg, the volume is 0.8 cubic meters, and the full load time is 4 p.m.; at 12 noon, the garbage weight increases to 150 kg, the volume is 1.2 cubic meters, and the full load time is updated to 6 p.m.

[0118] Using these historical data, the AI ​​prediction algorithm module will first build a linear regression prediction model. Model formula T = β 0 +β 1 *W+β 2 *Model parameter β in V+ε 0 , β 1 , β 2The solution will be obtained through optimization algorithms such as the least squares method to minimize the prediction error. Assume that the parameters obtained are β 0 =2,β 1 =0.01,β 2 =0.5, then the linear regression prediction model can be expressed as T=2+0.01W+0.5V+ε.

[0119] Subsequently, the AI ​​prediction algorithm module will continue to receive real-time data, such as the weight and volume of the garbage at the current time, and use this data to update and correct the linear regression prediction model. For example, if the current time is 2 pm, the weight of the garbage in the garbage bin is 180 kg and the volume is 1.5 cubic meters, the AI ​​prediction algorithm module will substitute this data into the linear regression prediction model for prediction, and the estimated full load time is 7 pm.

[0120] Next, the AI ​​prediction algorithm module will compare the prediction result with the preset time threshold T0. Assuming that the preset time threshold T0 is 6 pm, since the predicted full load time T (7 pm) is greater than the preset time threshold T0 (6 pm), there is no need to trigger the notification mechanism at this time. However, if the predicted full load time T is less than or equal to the preset time threshold T0, the AI ​​prediction algorithm module will determine that the notification mechanism needs to be triggered and generate the notification content according to the preset template.

[0121] In an embodiment of the present application, a storage medium stores a computer-readable program therein, and when the computer-readable program is called by a controller, the steps of the intelligent trash bin full load prediction and notification method as described in the present invention can be executed.

[0122] In an embodiment of the present application, the storage medium may be a tangible storage medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. The storage medium may be a machine-readable signal medium or a machine-readable storage medium. The storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0123] Embodiment 2

[0124] like Figure 2 As shown, in a possible embodiment, a method for predicting and notifying a full load of an intelligent garbage bin further includes:

[0125] Get the concentration of harmful gases in the garbage bin;

[0126] Obtain image information of garbage in the garbage bin, and identify the type of garbage based on the image information;

[0127] Determine the hazard level based on the concentration of harmful gases and the type of garbage;

[0128] Calculate a comprehensive score based on the hazard level and estimated full load time;

[0129] When the comprehensive score is greater than the preset score threshold, the notification mechanism is automatically triggered.

[0130] In one possible embodiment, the assessment of the garbage hazard level:

[0131] (1) Gas detection and analysis: Highly sensitive gas detection sensors, such as electrochemical sensors, PID photoionization detectors, etc., are installed inside the trash can to monitor the concentration of harmful gases in the trash can in real time, such as ammonia, hydrogen sulfide, etc. (corresponding sensors can also be configured as needed to detect sulfur dioxide, nitrogen compounds and volatile organic compounds). The presence of these gases reflects the degree of corruption of the garbage. In addition, through the camera installed in the trash can, combined with image recognition technology (such as convolutional neural network CNN in deep learning), the garbage can be classified and identified to determine whether there are preset dangerous types.

[0132] (2) Hazard level quantification: Combine the data collected by the gas detection sensor with the results of image recognition to quantify the hazard level according to the preset quantification standards. The specific quantification standards are as follows:

[0133] If the concentration of the harmful gas is greater than or equal to a first preset concentration threshold, the hazard level is high;

[0134] If the concentration of the harmful gas is less than the first preset concentration threshold and greater than or equal to the second preset concentration threshold, and the type of the garbage is a preset hazardous type, the hazard level is high, otherwise the hazard level is medium, wherein the first preset concentration threshold is greater than the second preset concentration threshold;

[0135] If the concentration of the harmful gas is less than the second preset concentration threshold, the hazard level is low.

[0136] In order to facilitate the quantitative comprehensive scoring later, the "hazard level" can be converted into a numerical value. The "high" hazard level corresponds to a numerical value of 3, the "medium" hazard level corresponds to a numerical value of 2, and the "low" hazard level corresponds to a numerical value of 1.

[0137] (3) Calculation of comprehensive score: The predicted full load time and hazard level are combined to form a comprehensive score system.

[0138] Specifically, the scoring can be done through the following comprehensive scoring formula:

[0139] Comprehensive score = α*(1 / estimated full load time)+β*hazard level;

[0140] In this formula, α and β are weight coefficients used to adjust the impact of full load time and hazard level on the comprehensive score.

[0141] For example, for a partially loaded trash bin, when the hazard level is high, a notification mechanism will be triggered even if the predicted full load time has not been reached so that it can be cleaned up first.

[0142] For example, in an intelligent waste management system, the ammonia concentration of trash bin 1 is monitored to be 0.09ppm and the hydrogen sulfide concentration is 0.05ppm through the high-sensitivity gas detection sensor installed in the trash bin, which exceeds the first preset concentration threshold (for example: ammonia concentration 0.05ppm, hydrogen sulfide concentration 0.03ppm), and the image recognition technology shows that there is a large amount of kitchen waste in the trash bin (assuming that the kitchen waste belongs to the preset dangerous category). Therefore, the hazard level of trash bin 1 is quantified as "high". If the ammonia concentration of trash bin 2 is 0.03ppm and the hydrogen sulfide concentration is 0.01ppm, both of which are lower than the second preset concentration threshold (for example: ammonia concentration 0.04ppm, hydrogen sulfide concentration 0.02ppm), and the image recognition result shows that the trash bin is mainly paper and a small amount of plastic waste (assuming that plastic waste does not belong to the preset dangerous category), its hazard level is quantified as "low". Assuming the full load prediction time, the expected full load time of trash bin 1 is 12 hours, and that of trash bin 2 is 24 hours.

[0143] According to the comprehensive scoring formula:

[0144] Comprehensive score = α*(1 / estimated full load time)+β*value corresponding to the hazard level);

[0145] Assume that in this embodiment, α and β are respectively set to 0.5, indicating that the full load time and the hazard level are equally important (the specific value can be flexibly adjusted).

[0146] The composite score for bin 1 is calculated as:

[0147] Comprehensive score = 0.5*(1 / 12)+0.5*3=0.5*0.0833+0.5*3=0.0417+1.5=1.5417;

[0148] The overall score of Trash Bin 2 is calculated as:

[0149] Comprehensive score = 0.5*(1 / 24)+0.5*1=0.5*0.0417+0.5*1=0.0208+0.5=0.5208;

[0150] Table 1: Comprehensive scoring results of trash bin 1 and trash bin 2:

[0151]

[0152] Assume that the preset score threshold is 1.5 (the preset score threshold is adjusted according to actual needs). Although trash bin 1 is not full at this time, its comprehensive score is greater than 1.5 points, so the notification mechanism is automatically triggered.

[0153] In a possible embodiment, after receiving the notification content, the notification module automatically notifies the cleaning personnel through SMS, APP push, email, etc. The cleaning personnel can go to the corresponding location in time to carry out garbage removal work according to the notification content.

[0154] When the notification module receives the notification content from the AI ​​prediction algorithm module, it will immediately start the notification mechanism and send the notification content to the cleaning staff in various ways. The notification module will send notifications to the cleaning staff via SMS. SMS notifications are highly instantaneous, ensuring that cleaning staff can quickly receive notifications and take appropriate measures. At the same time, SMS notifications also have the advantage of wide coverage, which can cover most cleaning staff. The notification module will also send notifications to cleaning staff through APP push, and the notification module will also send notifications to cleaning staff by email.

[0155] The notification content can be: [Smart Trash Bin Notification] Trash bin location: No. XX, XX Road, Trash bin No. 001 has a high hazard level, estimated full load time: 12 hours, please give priority to cleaning. Contact information: XXX-XXXX-XXXX.

[0156] like Figure 6 As shown, in a possible embodiment, a smart garbage bin full load prediction and notification system is divided according to functional modules, including an Internet of Things sensor module, a data acquisition and preprocessing module, an AI prediction algorithm module and a notification module, and also includes a comprehensive score calculation module. Among them, the comprehensive score calculation module is used to judge the hazard level based on the concentration of harmful gases and the type of garbage, and calculate the comprehensive score based on the hazard level and the expected full load time, and when the comprehensive score is greater than the preset score threshold, the notification mechanism is automatically triggered. The functions of the remaining modules are the same as those in Example 1.

[0157] In summary, an intelligent garbage bin full load prediction and notification system realizes real-time monitoring of garbage weight, volume change, harmful gas concentration and garbage type in the garbage bin through the close cooperation of multiple links such as IoT sensor module, data collection and preprocessing module, AI prediction algorithm module, comprehensive scoring calculation module and notification module, and determines whether to trigger the notification mechanism, and can promptly notify the cleaning staff to carry out garbage removal work. This system not only improves the efficiency and quality of garbage disposal, but also reduces the workload of cleaning staff, making a positive contribution to the beautification and protection of the urban environment.

[0158] The rest is the same as that of the first embodiment.

[0159] The above-mentioned embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above-mentioned embodiments. Any other changes, modifications, substitutions, combinations and simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement modes and shall be included in the protection scope of the present invention.

Claims

1. A method for predicting and notifying when a smart garbage bin is full, characterized in that: The following steps are involved: Obtain the weight and volume changes of garbage in the garbage bin; Perform preprocessing operations on the acquired data; Based on the historical and real-time data obtained, a linear regression prediction model is used to calculate the expected full load time of the trash bin and determine whether a notification mechanism needs to be triggered; wherein the linear regression prediction model is: T=β0+β1·W+β2·V+∈ Where T represents the expected full load time, W represents the weight of the garbage in the trash bin, V represents the volume of the garbage in the trash bin, β0, β1, and β2 are the parameters of the linear regression prediction model, which are learned through training data, and ∈ is the random error term; When it is determined that the trash bin is about to be full based on the estimated full load time, a notification mechanism is automatically triggered, where the content of the notification includes the specific location of the trash bin, the estimated full load time and cleaning suggestions.

2. The intelligent garbage bin full load prediction and notification method according to claim 1, characterized in that: When underload is determined based on the estimated full load time, it also includes: Obtain the concentration and image information of harmful gases in the garbage bin, and identify the type of garbage based on the image information; Determine the hazard level based on the concentration of harmful gases and the type of garbage; Calculate a comprehensive score based on the hazard level and estimated full load time; When the comprehensive score is greater than the preset score threshold, the notification mechanism is automatically triggered.

3. The intelligent garbage bin full load prediction and notification method according to claim 2, characterized in that: The specific method for determining the hazard level is: If the concentration of the harmful gas is greater than or equal to the first preset concentration threshold, the hazard level is high; If the concentration of the harmful gas is less than the first preset concentration threshold and greater than or equal to the second preset concentration threshold, and the type of the garbage is the preset hazardous type, the hazard level is high, otherwise the hazard level is medium, wherein the first preset concentration threshold is greater than the second preset concentration threshold; If the concentration of the harmful gas is less than the second preset concentration threshold, the hazard level is low.

4. The intelligent garbage bin full load prediction and notification method according to claim 2, characterized in that: The calculation formula of the comprehensive score is as follows: Comprehensive score = α*(1 / estimated full load time)+β*value corresponding to the hazard level); Among them, α represents the weight of the predicted full load time, and β represents the weight of the hazard level.

5. The intelligent garbage bin full load prediction and notification method according to claim 2, characterized in that: The harmful gas includes at least one of ammonia, hydrogen sulfide, sulfur dioxide, nitrogen-containing compounds and volatile organic compounds.

6. The intelligent garbage bin full load prediction and notification method according to claim 1, characterized in that: Perform preprocessing operations on the acquired data, specifically: Clean the data to remove invalid data and outliers; De-noising the data; Compress the data.

7. The intelligent garbage bin full load prediction and notification method according to claim 1, characterized in that: Based on the historical and real-time data obtained, a linear regression prediction model is used to calculate the expected full load time of the garbage bin, specifically: Build a linear regression prediction model that learns from historical data and understands the relationship between the weight and volume changes of garbage in the garbage bin and the full load time; After the linear regression prediction model is built, the linear regression prediction model is trained using historical data to solve the parameters β0, β1, and β2 of the linear regression prediction model to minimize the prediction error; Receive real-time data and use the real-time data to update and correct the linear regression prediction model; When the linear regression prediction model is updated, the estimated full load time of the garbage bin is calculated using the linear regression prediction model.

8. The intelligent garbage bin full load prediction and notification method according to claim 1, characterized in that: Notification is provided through at least one of SMS, APP push, and email.

9. An intelligent garbage bin full load prediction and notification system, characterized by: It comprises a memory and a controller, wherein the memory stores a computer-readable program, and when the computer-readable program is called by the controller, it can execute the steps of the intelligent garbage bin full load prediction and notification method as described in any one of claims 1 to 8.

10. A storage medium, characterized in that: A computer-readable program is stored therein, and when the computer-readable program is called by the controller, the steps of the intelligent garbage bin full load prediction and notification method as described in any one of claims 1 to 8 can be executed.

Images