A kind of urban hazardous waste monitoring and management system and emission prediction method

Abstract

The present invention discloses a city hazardous waste monitoring and management system and an emission prediction method, wherein the monitoring and management system comprises a hazardous waste centralized supervision platform, a cloud platform, an early warning unit, a hazardous waste collection unit, a hazardous waste transportation unit, and a hazardous waste treatment unit. The early warning unit accurately predicts the emission per unit time of different enterprises and different hazardous waste items through an emission prediction module, and the prediction method adopted by the emission prediction module comprises a weighted prediction method, and based on the difference between the actual collection amount and the predicted amount, more accurate management of different enterprises and different hazardous waste items and more accurate evaluation of the environmental protection work of enterprises are achieved.

Description

Technical Field

[0001] The present invention belongs to the field of hazardous waste monitoring and management, and in particular relates to an urban hazardous waste monitoring and management system and an emission prediction method. Background Art

[0002] Modern cities have many types of hazardous waste (hereinafter referred to as hazardous waste), involving many enterprises (hazardous waste emission sources), with a wide and scattered distribution, different emission amounts, and different emission times, which brings great difficulties to unified monitoring and collection management. With the efforts of the government and all sectors of society, relevant industries and enterprises have carried out a lot of research and are gradually forming a relatively complete and reasonable set of different hazardous waste management methods. For example, patents with announcement number CN111275211A titled “A solid waste information management system”, announcement number CN111291910A titled “CNOOC waste collection and management method”, and announcement number CN114066373A titled “Industrial waste collection and transfer system” are all methods for centralized collection and management of different types of hazardous wastes. Chinese patent applications with announcement number CN 103034178 A titled “Hazardous waste supervision system and electronic form management method” and CN104866982 B titled “A system and method for Internet + hazardous waste management” involve a system and method for hazardous waste management based on the Internet, which can realize intelligent management or information management and improve management efficiency, but the degree of intelligence is not yet high enough.

[0003] At the same time, due to the obvious characteristics of small, biased and scattered sources of hazardous waste in cities, it is difficult to effectively control behaviors such as discarding, directly dumping into sewers or public trash cans, and the actual effect of various systems on environmental protection needs to be further improved. Even if 24-hour monitoring is set up, it will not be very operational due to the high cost and workload of manual monitoring to detect violations in a timely manner. If the amount of hazardous waste recycling can be reasonably predicted, the actual collection amount can be compared and analyzed based on this, so that effective measures such as timely key inspections or urging enterprises to comply with rules and regulations can be taken to gradually realize the transition from institutional norms for environmental protection behaviors of enterprises or individuals to environmental protection behaviors becoming conscious behaviors of enterprises or individuals. Self-prediction by enterprises (referred to as self-testing) is a relatively easy method to think of, but its reliability and credibility need to be constantly verified. There is still a lack of a more reasonable prediction method for the emission per unit time (such as monthly or annually) for different enterprises and different hazardous waste products. Summary of the invention

[0004] In view of the above problems, the present invention designs a city hazardous waste monitoring and management system and an emission prediction method, and proposes an emission prediction module and a weighted prediction method, which not only improves the intelligence level of urban hazardous waste management, but also can achieve a more reasonable prediction of the emission per unit time of different enterprises and different hazardous waste products, thereby achieving more accurate management of urban hazardous waste.

[0005] The present invention designs a kind of urban hazardous waste monitoring and management system, including a cloud platform bidirectionally connected with a hazardous waste supervision platform, an early warning unit, a hazardous waste collection unit, a hazardous waste transportation unit and a hazardous waste treatment unit, the cloud platform is bidirectionally connected with the early warning unit, the hazardous waste collection unit, the hazardous waste transportation unit and the hazardous waste treatment unit, and is characterized in that the early warning unit includes an information transmission module, a prediction module, an early warning module and a data analysis module, and the prediction module is used to predict the emission per unit time of hazardous waste emission sources; the hazardous waste supervision platform includes an information transmission module, a display module, a planning module and an instruction module. The unit time can be a day, a week, a month, a season, a year, etc., which should be determined according to the size and stability cycle of various types of hazardous waste emissions. The daily emissions of most urban hazardous waste enterprises are small or obviously uneven, and it is generally advisable to select a monthly sampling interval. The annual emissions of a small number of enterprises or a certain type of hazardous waste are very small, and when the monthly emission is obviously uneven, it is advisable to adopt an annual sampling interval; the input information of the prediction module is mainly the sampling information preprocessed by the data analysis module.

[0006] Furthermore, the hazardous waste collection unit, hazardous waste transportation unit, and hazardous waste treatment unit all include information collection and transmission modules, and the information collection and transmission modules are used for local information collection, information transmission between units, and local alarms; the hazardous waste supervision platform, cloud platform, and early warning unit all include information transmission modules for information transmission between units; the information collection and transmission modules and information transmission modules transmit information through wired and / or wireless communication devices. The local alarm function of the information collection and transmission module includes alarms for whether the amount of hazardous waste collected during collection, transportation, and treatment is close to the upper limit of the hazardous waste collection container, whether the ambient temperature and humidity are close to or exceed the storage requirements of the hazardous waste, whether the power supply of the information collection and transmission module is close to the lower limit, whether the transportation process is abnormal, the amount of hazardous waste treatment completed, the amount of hazardous waste to be treated, and / or whether the storage time of hazardous waste is close to the upper limit. The alarm method includes sound and / or light. Each unit only alarms for local collection information, and at the same time receives alarm or early warning signals related to the unit sent by the hazardous waste supervision platform, which can drive the alarm synchronously with the local alarm signal.

[0007] Furthermore, at least one of a weight sensor, a liquid level sensor, a temperature sensor, a humidity sensor, a power sensor, a position sensor, a speed sensor, and a video sensor is provided in the information acquisition and transmission module, which is mainly used to collect monitoring information such as the weight, volume, ambient temperature, ambient humidity, power supply, vehicle position, vehicle speed, type of hazardous waste, hazardous waste treatment volume, and on-site video of hazardous waste, and complete information transmission through wired and / or wireless communication methods. The power supply refers to the power of an independent power supply that supplies power to the information acquisition and transmission module.

[0008] Furthermore, the hazardous waste management platform obtains information of each unit from the cloud platform, the planning module is used to prepare a work plan based on the emission prediction results of the prediction module, the warning information of the warning module and the real-time monitoring information of the hazardous waste collection unit, the instruction module is used to generate operation instructions and send them to at least one of the hazardous waste collection unit, the hazardous waste transportation unit and the hazardous waste treatment unit; the display module is used to display at least one of the real-time monitoring information, prediction information, work plan and operation instructions. The work plan is generally issued in advance based on the forecast results of the resources of each unit, including notifying the hazardous waste collection unit when and what kind of hazardous waste to collect, notifying the hazardous waste transportation unit when and where and which company to collect which hazardous waste, notifying the hazardous waste treatment unit when and which vehicle will transport which hazardous waste to the treatment unit, and notifying the hazardous waste treatment unit how much of which hazardous waste should be treated by when, etc.; operation instructions are generally issued by the hazardous waste supervision platform, including remote reset instructions for the information collection and transmission modules of each unit, status setting instructions, control instructions such as focus and rotation of the on-site video surveillance camera, real-time or quasi-real-time transportation instructions, real-time or quasi-real-time hazardous waste treatment instructions, etc. Real-time or quasi-real-time transportation instructions and treatment instructions are different from the plans issued in advance, and are temporary or emergency in nature.

[0009] Furthermore, the hazardous waste collection unit also includes a hazardous waste collection container, which includes at least one of a collection barrel, a collection box, a collection bag, and a collection bottle, and is used to collect gaseous, liquid, solid and / or colloidal hazardous wastes; the hazardous waste transportation unit also includes a transportation tool, which is used to complete the hazardous waste transportation work; the hazardous waste treatment unit also includes a treatment device, which is used to complete the treatment of hazardous wastes; the cloud platform also includes a database module, which is used to classify and store the information, warning information, work plans and / or operating instructions obtained by the information collection module. The database design is a prior art; generally, an information collection and transmission module needs to be set at an appropriate location or near the collection container, transportation tool, and treatment device to ensure that the information to be collected is timely and accurate.

[0010] Furthermore, the early warning unit obtains information from the cloud platform; the data analysis module is used to preprocess and analyze the monitoring information collected by the hazardous waste collection unit, the hazardous waste transportation unit, and the hazardous waste treatment unit; the early warning module is used to warn whether the collection amount during the hazardous waste collection, transportation, and treatment process exceeds the limit, and the early warning method includes sound and / or light method. The early warning information includes whether the collection amount during the hazardous waste collection, transportation, and treatment process is close to the upper limit of the hazardous waste collection container, whether the ambient temperature and humidity are close to or exceed the storage requirements of the hazardous waste, whether the power supply of the information collection and transmission module is close to the lower limit, whether the transportation process is abnormal, the amount of hazardous waste treatment completed, the amount of hazardous waste to be processed and / or whether the storage time of hazardous waste is close to the upper limit, etc. The data analysis module mainly performs preprocessing and analysis of the original sampling data, including the judgment, identification or elimination of wild values, drawing of monitoring curves, etc., to provide more accurate on-site information for the prediction module; the warning information of the early warning module is generally sent to the hazardous waste supervision platform to drive the buzzer to sound an alarm and / or display the alarm through the display module on a large screen or indicator light, generally using a flashing red light alarm; the hazardous waste supervision platform can realize the alarm and display of all other units. The alarm information of each unit itself is generally collected and compared by the local information, and then the information collection and transmission module automatically issues an audible and visual alarm signal. The remote hazardous waste supervision platform also sends relevant audible and visual alarm signals to each unit. The specific alarm technology is the existing technology.

[0011] According to a hazardous waste emission prediction method for an urban hazardous waste monitoring and management system, the emission prediction method is implemented through a prediction module, which is used to accurately predict the emission per unit time of different hazardous waste emission sources and different hazardous waste items, including a weighted prediction method.

[0012] Furthermore, the weighted prediction method includes the following method:

[0013] Note x i is the predicted value of total waste discharge per unit time by the ith independent prediction method, w i is the i-th predicted value x i The weighting coefficient of , i∈{1, 2, ..., n}, n is the total number of independent prediction methods, then the weighted predicted total waste emissions per unit time x is:

[0014]

[0015] in,

[0016] When the reliability of each weighted coefficient is uncertain, the average value w can be directly taken. i =1 / n; when a certain independent prediction method is more accurate, its weight coefficient can be increased; conversely, when a certain independent prediction method is obviously uncertain, its weight coefficient can be reduced until it reaches 0.

[0017] The independent prediction method includes at least one of a prediction method based on economic indicators, a prediction method based on long-term monitoring data and a filtering model, a self-test method of an emission source entity, and a prediction method based on relevant import and export indicators. Prediction based on economic indicators generally refers to calculating theoretical emissions using the impact coefficient of relevant hazardous waste emissions on total turnover and / or tax revenue; prediction based on long-term monitoring data and a filtering model refers to recording emission data at fixed sampling intervals and using existing filtering models, such as Kalman filtering, for prediction; enterprise self-test generally refers to an estimate of emissions per unit time given by an enterprise based on its own experience data; prediction based on relevant import and export indicators refers to a certain type of emissions from A being closely related to the amount of goods used by B, and the amount of goods used can be determined based on the purchase amount and inventory, so that the emissions of A can be calculated based on the amount of goods used by B.

[0018] Furthermore, the weighted prediction method also includes the following method:

[0019] Let x1 be the predicted value of total waste discharge per unit time based on economic indicators, then

[0020] x1=(a1b1+a2b2) / 2 Formula (2)

[0021] Among them, a1 is the tax impact coefficient, b1 is the monthly total tax, a2 is the turnover impact coefficient, and b2 is the monthly total turnover;

[0022] x2 is the predicted value of the total waste discharge per unit time based on long-term monitoring data and a filtering model, wherein the filtering model includes a state model and a filtering method, wherein the state model includes a constant velocity model, a constant acceleration model, a polynomial model or a current model, etc.; and the filtering method includes a Kalman filtering method;

[0023] x3 is the total waste emission per unit time value measured by the enterprise itself, and the predicted value x of the total waste emission per unit time output by the emission prediction module is:

[0024] x=x1w1+x2w2+x3w3 Formula (3)

[0025] Wherein w1, w2, w3 are weighted coefficients of x1, x2, x3 respectively, and w1+w2+w3=1.

[0026] Furthermore, the filtering model includes a Kalman filtering method based on a constant velocity model, and its iterative steps are:

[0027] Let k = 0, 1, 2, 3, ..., the measured waste discharge amount z of a certain type at time k , whose state estimate is x k , the unit time rate of change of its state estimate is State Vector State transition matrix t is the sampling unit time, and the covariance matrix is ​​P k , the measurement matrix H k =

[10] ;

[0028] Take the initialization time k = 0 to monitor the waste discharge z0 in real time, which is the state quantity x0, that is, x0 = z0, then initialize Pick Then perform filtering and prediction according to the following steps:

[0029] S1: One-step prediction of the state from time k to time k+1

[0030] X k+1_k =F k X k Formula (4)

[0031] S2: state one-step forecast covariance

[0032]

[0033] The superscript T indicates the matrix transpose, Q k is the state noise matrix, which is determined according to the noise of the state quantity;

[0034] S3: Kalman gain matrix

[0035]

[0036] Among them, R k is the measurement noise matrix at time k, which is determined based on the noise of the actual measurement value;

[0037] S4: Status Update

[0038] X k+1 =X k +K k+1 (z k+1 -H k+1 X k+1_k ) Formula (7)

[0039] S5: State update covariance

[0040] P k+1 =(IK k+1 H k+1 ) k+1_k Formula (8)

[0041] Where I is the identity matrix;

[0042] Steps S1 to S5 are executed repeatedly to obtain the state vector at any sampling time k. The predicted state vector at time k+1 is obtained by formula (4). The first element of the predicted state vector is the filter model predicted value of the total waste discharge at time k+1.

[0043] The advantages and beneficial effects of the present invention are as follows: the urban hazardous waste monitoring and management system designed by the present invention realizes the intelligent centralized collection and treatment of hazardous waste emissions from small, remote and scattered enterprises in every corner of the city, especially realizes the reasonable prediction of the unit time emissions of different enterprises and different hazardous waste products, can timely discover unreasonable phenomena, strengthen inspection and supervision through administrative law enforcement, focus on reviewing monitoring video information, etc., so as to effectively control the private or arbitrary emissions of enterprises, improve the effectiveness of the urban hazardous waste monitoring and management system, and improve the level and ability of environmental protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 The present invention is a block diagram of a city hazardous waste monitoring and management system. DETAILED DESCRIPTION

[0045] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings and embodiments. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0046] Example 1: A city hazardous waste monitoring and management system

[0047] The present invention provides a city hazardous waste monitoring and management system, such as Figure 1As shown, it includes a cloud platform bidirectionally connected to the hazardous waste supervision platform, an early warning unit, a hazardous waste collection unit, a hazardous waste transportation unit and a hazardous waste treatment unit. The cloud platform is bidirectionally connected to the early warning unit, the hazardous waste collection unit, the hazardous waste transportation unit and the hazardous waste treatment unit. It is characterized in that the early warning unit includes an information transmission module, a prediction module, an early warning module and a data analysis module. The prediction module is used to predict the emission per unit time of the hazardous waste emission source; the hazardous waste supervision platform includes an information transmission module, a display module, a planning module and an instruction module. The unit time can be day, week, month, season, year, etc., which should be determined according to the size and stable cycle of various types of hazardous waste emissions. The daily emissions of most urban hazardous waste enterprises are small or obviously uneven. Generally, it is advisable to select a monthly sampling interval. The annual emissions of a small number of enterprises or a certain type of hazardous waste are very small, and the annual sampling interval is suitable when the monthly emission is obviously uneven; the input information of the prediction module is mainly the sampling information preprocessed by the data analysis module, and the data analysis module directly obtains the original sampling information from the information transmission module. Based on the existing technology, the present invention adds a data analysis module and a prediction module to the early warning unit, providing a more reasonable and accurate prediction of hazardous waste emissions, thereby judging whether the enterprise has fully or basically complied with the hazardous waste collection and emission regulations, and providing a basis for administrative law enforcement agencies to determine key inspection targets; independent planning modules and instruction modules are designed for the hazardous waste supervision platform, which together with the improvement of the early warning unit improve the intelligence level of the system.

[0048] Preferably, the hazardous waste collection unit, hazardous waste transportation unit, and hazardous waste treatment unit all include an information acquisition and transmission module, and the information acquisition and transmission module is used for local information acquisition, information transmission between units, and local alarm; the hazardous waste supervision platform, cloud platform, and early warning unit all include an information transmission module, which is used for information transmission between units; the information acquisition and transmission module and the information transmission module transmit information through wired and / or wireless communication devices. The local alarm function of the information collection and transmission module includes alarming whether the collection amount during the hazardous waste collection, transportation and treatment process is close to the upper limit of the hazardous waste collection container, whether the ambient temperature and humidity are close to or exceed the storage requirements of the hazardous waste, whether the power supply of the information collection and transmission module is close to the lower limit, whether the transportation process is abnormal, the amount of hazardous waste treatment completed, the amount of hazardous waste to be treated and / or whether the storage time of hazardous waste is close to the upper limit. The alarm method includes sound and / or light method. Each unit only alarms for locally collected information, and at the same time receives alarm or early warning signals related to the unit sent by the hazardous waste supervision platform, which can drive the alarm synchronously with the local alarm signal; the hazardous waste transportation unit in this embodiment adopts a 4G / 5G adaptive wireless network to connect to the hazardous waste supervision platform and the cloud platform, and information transmission between other units directly uses the Internet connection.

[0049] Preferably, at least one of a weight sensor, a liquid level sensor, a temperature sensor, a humidity sensor, a power sensor, a position sensor, a speed sensor, and a video sensor is provided in the information acquisition and transmission module, which is mainly used to collect monitoring information such as the weight, volume, ambient temperature, ambient humidity, power supply, vehicle position, vehicle speed, type of hazardous waste carried, hazardous waste treatment volume, and on-site video of hazardous waste, and complete information transmission through wired and / or wireless communication methods. Power supply refers to the power of an independent power supply that supplies power to the information acquisition and transmission module. Power supply refers to the power of an independent power supply that supplies power to the information acquisition and transmission module. The hazardous waste collection unit of this embodiment mainly collects information such as hazardous waste weight, volume, ambient temperature, ambient humidity, power supply and on-site video. The corresponding enterprise name and hazardous waste type information are pre-entered into the database and bound to the hazardous waste collection container; the hazardous waste transportation unit mainly collects information such as power supply, vehicle position, vehicle speed, hazardous waste type, hazardous waste weight, ambient temperature, ambient humidity, etc. The hazardous waste type information, vehicle type, license plate and other information are pre-entered into the database and bound to the planned task; the hazardous waste treatment unit mainly collects information such as power supply, various types of hazardous waste storage, treatment volume, inventory, storage environment temperature, storage environment humidity, on-site video, etc. More detailed information is generally collected and processed by the hazardous waste treatment plant; the time information corresponding to various types of sampling information is automatically acquired by the system; the installation of sensors is flexibly determined according to the on-site conditions. Sensors that allow conditions are integrated with the information collection and transmission module body. Sensors that do not allow conditions are installed in appropriate locations and communicate with the information collection and transmission module body through signal cables or wireless networks. Regardless of the installation method, this embodiment regards it as a part of the information collection and transmission module.

[0050] Preferably, the hazardous waste supervision platform obtains information of each unit from the cloud platform, the planning module is used to prepare a work plan in combination with the emission prediction results of the prediction module, the warning information of the warning module and the real-time monitoring information of the hazardous waste collection unit, the instruction module is used to generate operation instructions and send them to at least one of the hazardous waste collection unit, the hazardous waste transportation unit and the hazardous waste treatment unit; the display module is used to display at least one of the real-time monitoring information, the prediction information, the work plan and the operation instructions. The work plan is generally issued in advance according to the resource situation of each unit according to the prediction results, including notifying the hazardous waste collection unit when to collect what kind of hazardous waste, notifying the hazardous waste transportation unit when to collect what kind of hazardous waste, notifying the hazardous waste treatment unit when and which vehicle will transport what kind of hazardous waste to the treatment unit, notifying the hazardous waste treatment unit how much and what kind of hazardous waste should be processed by when, etc.; the operation instructions are generally issued by the hazardous waste supervision platform, including remote reset instructions for the information collection and transmission modules of each unit, status setting instructions, control instructions such as camera focus and rotation of on-site video monitoring, real-time or quasi-real-time transportation instructions, real-time or quasi-real-time hazardous waste treatment instructions, etc., real-time or quasi-real-time transportation instructions, real-time or quasi-real-time hazardous waste treatment instructions, etc. Instructions and processing instructions are different from the plans issued in advance in nature, and are temporary or emergency in nature. They can generally be issued one day in advance or executed immediately. In this embodiment, the normal work plan is issued one week in advance, that is, the full work plan for the following week is issued on Monday. For enterprises with smaller hazardous waste emissions, the collection plan is issued in the form of one week's advance notice, and it does not need to be issued at a fixed cycle. The information collected by each information collection unit is directly transmitted to the cloud platform for storage, and other units obtain new information through the cloud platform. Under special circumstances, the hazardous waste supervision platform can issue instructions and can also communicate directly with each unit through the setting of the information transmission module, including directly obtaining on-site collection information, but it does not affect the active transmission of each collected information to the cloud platform.

[0051] Preferably, the hazardous waste collection unit also includes a hazardous waste collection container, and the collection container includes at least one of a collection barrel, a collection box, a collection bag, and a collection bottle, which is used to collect gaseous, liquid, solid and / or colloidal hazardous waste; the hazardous waste transportation unit also includes a transportation tool for completing the hazardous waste transportation work; the hazardous waste treatment unit also includes a treatment device for completing the treatment of hazardous waste; the cloud platform also includes a database module for classifying and storing the information, warning information, work plans and / or operating instructions obtained by the information collection module. The database design is a prior art; generally, an information collection and transmission module needs to be set at an appropriate location or near the collection container, transportation tool, and treatment device to ensure that the information to be collected is timely and accurate. This embodiment integrates the information collection and transmission module with the collection container, transportation tool, and treatment device to ensure that the information to be collected is timely and accurate. In the engineering application of this embodiment, the hazardous waste discharge behavior of the hazardous waste collection unit needs to be carried out under the centralized supervision of the hazardous waste supervision platform; the hazardous waste transportation unit can operate independently, but its entire transportation process and plan must be subject to the centralized supervision of the hazardous waste supervision platform; the hazardous waste treatment unit works relatively independently and generally operates independently, mainly feedbacks the overall treatment situation to the hazardous waste supervision platform and receives guiding treatment plans, and the specific treatment work unit is responsible for it.

[0052] Preferably, the early warning unit obtains information from the cloud platform; the data analysis module is used to preprocess and analyze the monitoring information collected by the hazardous waste collection unit, the hazardous waste transportation unit, and the hazardous waste treatment unit; the early warning module is used to warn whether the collection amount during the hazardous waste collection, transportation, and treatment process exceeds the limit, and the early warning method includes sound and / or light method. The early warning information includes whether the collection amount during the hazardous waste collection, transportation, and treatment process is close to the upper limit of the hazardous waste collection container, whether the ambient temperature and humidity are close to or exceed the storage requirement of the hazardous waste, whether the power supply of the information collection and transmission module is close to the lower limit, whether the transportation process is abnormal, the amount of hazardous waste treatment completed, the amount of hazardous waste to be processed, and / or whether the storage time of hazardous waste is close to the upper limit, etc. The data analysis module mainly performs preprocessing and analysis of the original sampling data, including the judgment, identification or elimination of wild values, drawing of monitoring curves, etc., to provide more accurate on-site information for the prediction module; the warning information of the early warning module is generally sent to the hazardous waste supervision platform to drive the buzzer to sound an alarm and / or display the alarm through the display module on a large screen or indicator light, generally using a flashing red light alarm; the hazardous waste supervision platform can realize the alarm and display of all other units. The alarm information of each unit itself is generally collected and compared by the local information, and then the information collection and transmission module automatically issues an audible and visual alarm signal. The remote hazardous waste supervision platform also sends relevant audible and visual alarm signals to each unit. The specific alarm technology is the existing technology.

[0053] Example 2: A method for predicting hazardous waste emissions

[0054] According to a hazardous waste emission prediction method for an urban hazardous waste monitoring and management system described in Example 1, the emission prediction method is implemented through a prediction module, which is used to accurately predict the emission per unit time of different hazardous waste emission sources and different hazardous waste items, including a weighted prediction method.

[0055] Preferably, the weighted prediction method includes the following method:

[0056] Note x i is the predicted value of total waste discharge per unit time by the ith independent prediction method, w i is the i-th predicted value x i The weighting coefficient of , i∈{1, 2, ..., n}, n is the total number of independent prediction methods, then the weighted predicted total waste emissions per unit time x is:

[0057]

[0058] in,

[0059] When the reliability of each weighted coefficient is uncertain, the average value w can be directly taken. i =1 / n; when a certain independent prediction method is more accurate, its weight coefficient can be increased; conversely, when a certain independent prediction method is obviously uncertain, its weight coefficient can be reduced until it is 0. In this embodiment, the weight coefficient takes the average value.

[0060] The independent prediction method includes at least one of a prediction method based on economic indicators, a prediction method based on long-term monitoring data and a filtering model, a self-test method of an emission source entity, and a prediction method based on relevant import and export indicators. Prediction based on economic indicators generally refers to calculating theoretical emissions using the impact coefficient of relevant hazardous waste emissions on total turnover and / or tax revenue; prediction based on long-term monitoring data and a filtering model refers to recording emission data at fixed sampling intervals and using existing filtering models, such as Kalman filtering, for prediction; enterprise self-test generally refers to an estimate of emissions per unit time given by an enterprise based on its own experience data; prediction based on relevant import and export indicators refers to a certain type of emissions from A being closely related to the amount of goods used by B, and the amount of goods used can be determined based on the purchase amount and inventory, so that the emissions of A can be calculated based on the amount of goods used by B.

[0061] Example 3: Another method for predicting hazardous waste emissions

[0062] According to another hazardous waste emission prediction method of an urban hazardous waste monitoring and management system described in Example 1, the emission prediction method is implemented by a prediction module, which is used to accurately predict the emission per unit time of different hazardous waste emission sources and different hazardous waste items, including a weighted prediction method.

[0063] Preferably, the weighted prediction method includes the following method:

[0064] Let x1 be the predicted value of total waste discharge per unit time based on economic indicators, then

[0065] x1=(a1b1+a2b2) / 2 Formula (2)

[0066] Among them, a1 is the tax impact coefficient, b1 is the monthly total tax, a2 is the turnover impact coefficient, and b2 is the monthly total turnover; if 1kg of emissions should increase turnover by 2,000 yuan, accounting for about 1 / 2 of the total turnover, and increase business tax by 120 yuan, accounting for about 1 / 3 of the total tax, then

[0067] x2 is the predicted value of the total waste discharge per unit time based on long-term monitoring data and a filtering model, wherein the filtering model includes a state model and a filtering method, wherein the state model includes a constant velocity model, a constant acceleration model, a polynomial model or a current model, etc.; and the filtering method includes a Kalman filtering method;

[0068] x3 is the total waste emission per unit time value measured by the enterprise itself, and the predicted value x of the total waste emission per unit time output by the emission prediction module is:

[0069] x=x1w1+x2w2+x3w3 Formula (3)

[0070] Wherein w1, w2, w3 are weighted coefficients of x1, x2, x3 respectively, and w1+w2+w3=1. In this embodiment, w1=w2=w3=1 / 3 is taken.

[0071] Preferably, the filtering model described in Embodiments 2 and 3 includes a Kalman filtering method based on a constant velocity model, and the iterative steps are:

[0072] Let k = 0, 1, 2, 3, ..., the measured waste discharge amount z of a certain type at time k , whose state estimate is x k , the unit time rate of change of its state estimate is State Vector State transition matrix t is the sampling unit time, and the covariance matrix is ​​P k , the measurement matrix H k =

[10] ;

[0073] Take the initialization time k = 0 to monitor the waste discharge z0 in real time, which is the state quantity x0, that is, x0 = z0, then initialize Pick Then perform filtering and prediction according to the following steps:

[0074] S1: One-step prediction of the state from time k to time k+1

[0075] X k+1_k =F k X k Formula (4)

[0076] S2: state one-step forecast covariance

[0077]

[0078] The superscript T indicates the matrix transpose, Q k is the state noise matrix, which is determined according to the noise of the state quantity; Q k For adjustable data, take the waste oil of a car repair station as an example, in kg, it can be obtained

[0079] S3: Kalman gain matrix

[0080]

[0081] Among them, R k is the measurement noise matrix at time k, which is determined according to the noise of the actual measurement value; R k For adjustable data, take the waste oil of a car repair station as an example, in kg, we can take R k =5;

[0082] S4: Status Update

[0083] X k+1 =X k +K k+1 (z k+1 -H k+1 X k+1_k ) Formula (7)

[0084] S5: State update covariance

[0085] P k+1 =(IK k+1 H k+1 ) k+1_k Formula (8)

[0086] Where I is the identity matrix;

[0087] Steps S1 to S5 are executed repeatedly to obtain the state vector at any sampling time k. The predicted state vector at time k+1 is obtained by formula (4). The first element of the predicted state vector is the filter model predicted value of the total waste discharge at time k+1.

[0088] The sampling time interval can be based on the actual hazardous waste collected, and can generally be seconds, minutes, hours, days, weeks, months or years. Since daily emissions are generally more random, hazardous waste emissions are usually sampled at monthly intervals. If the enterprise has daily emissions and the randomness of the emissions is relatively small, it is recommended to use daily or weekly intervals. Since the annual interval is relatively long, it is not conducive to filter initialization and is generally not recommended. Only when the annual emissions are small and the monthly emissions are extremely uneven, the annual interval sampling time is considered.

[0089] The basic principle of the present invention is to apply modern Internet of Things sensor technology, network interconnection technology and filtering prediction technology to urban hazardous waste monitoring and management. This is a research and development direction worth trying and working hard on, and it is of great significance to environmental protection and healthy development. Due to the small, biased and scattered characteristics of urban hazardous waste emission enterprises and the pursuit of profits and convenience by enterprises, it is difficult to effectively and consciously collect hazardous waste. Therefore, a more intelligent hazardous waste monitoring and management system is established, and the non-environmentally friendly behavior of enterprises can be effectively controlled by reasonably predicting the emission volume and comparing it with the actual centralized collection volume. Since there are many factors related to the emission of hazardous waste by enterprises, but it is difficult to correspond one by one, a variety of independent prediction methods can be generated, and various independent prediction methods always have their uncertainties. According to statistical principles, statistical weighting of multiple uncertain results can obtain more reliable results. Therefore, the present invention proposes a weighted prediction method, which can achieve more accurate management of different enterprises and different hazardous waste materials, and more accurate evaluation of the environmental protection work of enterprises. In fact, if n=1, only the self-test results of the enterprise are used as the emission prediction results, and the weighting coefficient is also 1, which is the simplest emission prediction method that can usually be thought of. On the other hand, the present invention focuses on further improving the intelligent level of urban hazardous waste management through the reasonable design and application of prediction modules, early warning modules, data analysis modules, instruction modules and planning modules.

[0090] In summary, the present invention provides a city hazardous waste monitoring and management system, which can realize intelligent centralized collection and treatment of city hazardous waste, effectively control the private or arbitrary discharge behavior of enterprises, improve the effectiveness of the city hazardous waste monitoring and management system, and improve the level and ability of environmental protection. The focus of the invention is to achieve more reasonable prediction of the emission of different enterprises and different hazardous waste items. Various weighting coefficients can be determined based on experience, and can also be adjusted based on actual use and inspection results. The filtering model can select a more reasonable model based on the nature and change law of the monitored quantity. The embodiment only provides one implementation method.

[0091] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention, including using different weighting methods or filtering models to derive different prediction methods. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A city hazardous waste monitoring and management system, comprising a cloud platform bidirectionally connected to a hazardous waste supervision platform, an early warning unit, a hazardous waste collection unit, a hazardous waste transportation unit and a hazardous waste treatment unit, wherein the cloud platform is bidirectionally connected to the early warning unit, the hazardous waste collection unit, the hazardous waste transportation unit and the hazardous waste treatment unit, and characterized in that: The early warning unit includes an information transmission module, a prediction module, an early warning module and a data analysis module. The prediction module is used to predict the emission per unit time of hazardous waste emission sources; the hazardous waste supervision platform includes an information transmission module, a display module, a planning module and an instruction module; Also included is a hazardous waste emission prediction method, which is implemented through a prediction module and is used to accurately predict the emission per unit time of different hazardous waste emission sources and different hazardous waste items, including a weighted prediction method; The weighted prediction method includes the following methods: Note x i is the predicted value of total waste discharge per unit time by the ith independent prediction method, w i is the i-th predicted value x i The weighting coefficient of , i∈{1, 2, ..., n}, n is the total number of independent prediction methods, then the weighted predicted total waste emissions per unit time x is: in, The independent prediction method includes at least one of a prediction method based on economic indicators, a prediction method based on long-term monitoring data and a filtering model, a self-test method of an emission source entity, and a prediction method based on relevant import and export indicators; The weighted prediction method comprises: Let x1 be the predicted value of total waste discharge per unit time based on economic indicators, then x1=(a1b1+a2b2) / 2 Formula (2) Among them, a1 is the tax impact coefficient, b1 is the monthly total tax, a2 is the turnover impact coefficient, and b2 is the monthly total turnover; x2 is the predicted value of the total waste discharge per unit time based on long-term monitoring data and a filtering model, wherein the filtering model includes a state model and a filtering method, wherein the state model includes a constant velocity model, a constant acceleration model, a polynomial model or a current model; and the filtering method includes a Kalman filtering method; x3 is the total waste emission per unit time value measured by the emission source, and the predicted value x of the total waste emission per unit time output by the emission prediction module is: x=x1w1+x2w2+x3w3 Formula (3) Wherein w1, w2, w3 are weighted coefficients of x1, x2, x3 respectively, and w1+w2+w3=1.

2. The urban hazardous waste monitoring and management system according to claim 1 is characterized in that: The hazardous waste collection unit, hazardous waste transportation unit, and hazardous waste treatment unit all include an information acquisition and transmission module, which is used for local information acquisition, information transmission between units, and local alarm; the hazardous waste supervision platform, cloud platform, and early warning unit all include an information transmission module, which is used for information transmission between units; the information acquisition and transmission module and the information transmission module transmit information through wired and / or wireless communication devices.

3. The urban hazardous waste monitoring and management system according to claim 2 is characterized in that: The information collection and transmission module is provided with at least one of a weight sensor, a liquid level sensor, a temperature sensor, a humidity sensor, a power sensor, a position sensor, a speed sensor, and a video sensor.

4. The urban hazardous waste monitoring and management system according to claim 1 is characterized in that: The hazardous waste supervision platform obtains information of each unit from the cloud platform. The planning module is used to prepare a work plan based on the emission prediction results of the prediction module, the warning information of the warning module and the real-time monitoring information of the hazardous waste collection unit. The instruction module is used to generate operation instructions and send them to at least one of the hazardous waste collection unit, the hazardous waste transportation unit, and the hazardous waste treatment unit; the display module is used to display at least one of the real-time monitoring information, prediction information, work plan, and operation instructions.

5. The urban hazardous waste monitoring and management system according to claim 1 is characterized in that: The hazardous waste collection unit also includes a hazardous waste collection container, which includes at least one of a collection barrel, a collection box, a collection bag, and a collection bottle; the hazardous waste transportation unit also includes a transportation tool for completing the hazardous waste transportation work; the hazardous waste treatment unit also includes a treatment device for completing the treatment of hazardous waste; the cloud platform also includes a database module for classifying and storing the information, early warning information, work plans and / or operating instructions obtained by the information collection module.

6. An urban hazardous waste monitoring and management system according to any one of claims 1 to 5, characterized in that: The early warning unit obtains information from the cloud platform; the data analysis module is used to preprocess and analyze the monitoring information collected by the hazardous waste collection unit, the hazardous waste transportation unit, and the hazardous waste treatment unit; the early warning module is used to warn whether the collection volume during the hazardous waste collection, transportation, and treatment process exceeds the limit, and the early warning method includes sound and / or light.

7. A method for predicting hazardous waste emissions according to claim 1, characterized in that: The filtering model includes a Kalman filtering method based on a constant velocity model, and its iterative steps are: Let k = 0, 1, 2, 3, ..., the measured waste discharge amount z of a certain type at time k , whose state estimate is x k , the unit time rate of change of its state estimate is State Vector State transition matrix t is the sampling unit time, and the covariance matrix is ​​P k , the measurement matrix H k =[10]; Take the initialization time k = 0 to monitor the waste discharge z0 in real time, which is the state quantity x0, that is, x0 = z0, then initialize Pick Then perform filtering and prediction according to the following steps: S1: One-step prediction of the state from time k to time k+1 X k+1_k =F k X k Formula (4) S2: state one-step forecast covariance The superscript T indicates the matrix transpose, Q k is the state noise matrix, which is determined according to the noise of the state quantity; S3: Kalman gain matrix K k+1 Among them, R k is the measurement noise matrix at time k, which is determined based on the noise of the actual measurement value; S4: Status Update X k+1 =X k +K k+1 (z k+1 -H k+1 X k+1_k ) Formula (7) S5: State update covariance P k+1 =(IK k+1 H k+1 ) k+1_k Formula (8) Where I is the identity matrix; Steps S1 to S5 are executed repeatedly to obtain the state vector at any sampling time k. The predicted state vector at time k+1 is obtained by formula (4). The first element of the predicted state vector is the filter model predicted value of the total waste discharge at time k+1.

Images