Multi-party intelligent supervision system for muck

Abstract

The invention discloses a multi-party intelligent monitoring system for residual mud and muck, which comprises a data acquisition module, a data processing module and a multi-party monitoring platform, and is characterized in that the data processing module is used for acquiring the mud output, the number of transportation times, the transportation path, the scattering and leaking condition and the soil scheduling amount according to the acquired data; and sending the mud discharge amount, the number of times of transportation, the transportation path, the scattering and leaking conditions and the soil amount scheduling amount to a multi-party supervision platform. The method can significantly improve the supervision efficiency, improves the environmental benefits, promotes the cyclic utilization of resources, and increases the public participation degree.

Description

Technical Field

[0001] The present invention relates to the technical field of slag management, and more specifically to a multi-party intelligent supervision system for residual mud and slag. Background Art

[0003] At present, the supervision of urban muck trucks mainly relies on manual inspections, which makes it difficult to achieve real-time monitoring and detect problems in a timely manner. At the same time, the existing systems often lack data analysis and prediction capabilities, and cannot quickly adjust management strategies based on real-time data. In addition, public participation is low, information is not open, and it is difficult for the public to obtain relevant data, which affects the public's supervision and participation in earthwork management. The existing complaint and feedback mechanism is also not perfect, resulting in poor public feedback channels.

[0004] In addition, the main method of dealing with residual mud and debris in the past was landfill, which not only wasted precious soil resources and building materials, but also failed to realize their potential value.

[0005] Therefore, there is an urgent need for a multi-party intelligent supervision system for residual mud and slag to achieve effective supervision of dump trucks, improve management efficiency, protect the environment, and promote the recycling of residual mud and slag. Summary of the invention

[0006] In view of this, in order to at least partially solve the above-mentioned technical problems, the present invention provides a multi-party intelligent supervision system for residual mud and slag.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A multi-party intelligent supervision system for residual mud and slag includes a data acquisition module, a data processing module and a multi-party supervision platform. The data processing module is used to obtain the mud output, transportation times, transportation routes, leakage conditions and soil scheduling quantity according to the collected data; and send the mud output, transportation times, transportation routes, leakage conditions and soil scheduling quantity to the multi-party supervision platform.

[0009] Furthermore, the data acquisition module includes a planning data acquisition unit, a construction site data acquisition unit, a market transaction data acquisition unit, a traffic video data acquisition unit, a multi-scene sensor data acquisition unit and an environmental weather data acquisition unit.

[0010] Furthermore, the planning data collection unit is used to collect topographic and geological data, construction design data, environmental protection and regulatory data, and surrounding planning data;

[0011] The site data collection unit is used to collect excavation and filling data, soil and geological data, construction progress and stage data, transportation and processing data;

[0012] The market transaction data collection unit is used to collect slag treatment cost data, resource recovery and reuse data, supply and demand relationship data, policy and regulatory data;

[0013] The traffic video data acquisition unit is used to collect road traffic flow, traffic congestion, public transportation routes, traffic checkpoint video data, load monitoring data, transportation route tracking data and safety compliance data;

[0014] The multi-scenario sensor data acquisition unit is used to collect construction site earthwork data, construction site exit measurement data, disposal site entry measurement data, construction site and disposal site environmental monitoring data, and transport vehicle tracking data;

[0015] The environmental weather data collection unit is used to collect precipitation data, temperature and humidity data, wind speed and wind direction data.

[0016] Furthermore, the multi-party supervision platform includes a market supervision platform, a government supervision platform and a public supervision platform.

[0017] Furthermore, the mud output, number of transportation times, transportation route, and leakage situation are sent to the government supervision platform; the transportation route is sent to the public supervision platform; the transportation route and soil scheduling amount are sent to the market supervision platform.

[0018] Furthermore, a neural network model is constructed to assist the government supervision platform in supervision. The neural network model is used to predict the amount of slag generated in each area or the amount generated by the slag truck based on the residual mud and slag distribution map and the residual mud and slag truck bucket, and to predict the future generation based on the historical earthwork volume and weather data.

[0019] It can be seen from the above technical solutions that the present invention discloses a multi-party intelligent supervision system for residual mud and slag, which can significantly improve supervision efficiency, improve environmental benefits, promote resource recycling and increase public participation compared with the existing technology; specifically, it includes:

[0020] Real-time monitoring and intelligent analysis: The supervision system of the present invention adopts advanced information technology and Internet of Things equipment, which can realize real-time monitoring of muck trucks and timely discover illegal behaviors; at the same time, the system has intelligent data analysis capabilities, which can perform real-time analysis and prediction based on the collected data, providing scientific and effective decision-making basis for management departments;

[0021] Multi-party participation and transparent management: The system of the present invention encourages public participation, provides information query and feedback channels, and increases regulatory transparency. By building a multi-party participation platform, the government, enterprises and the public can jointly participate in the supervision of muck trucks, forming an effective social supervision mechanism;

[0022] Improve supervision efficiency: Through automated and intelligent supervision methods, reduce reliance on manual inspections, improve supervision efficiency and accuracy, and reduce management costs;

[0023] Environmental protection and resource recycling: The system of the present invention promotes the recycling of residual mud and slag, reduces the damage to the environment caused by landfill, saves soil resources and building materials, and realizes green and sustainable development;

[0024] Data integration and sharing: The system integrates data from different sources to achieve information sharing, facilitate the collaboration of relevant departments and improve the overall management level;

[0025] Regulate the market order: through effective supervision, promote the standardization of the slag transportation market, reduce overloading, over-limit, illegal driving and other behaviors, and ensure road traffic safety;

[0026] Improve public satisfaction: By improving the management of garbage trucks, reducing environmental pollution and traffic interference, improving the quality of life of residents, and enhancing public satisfaction with urban management. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0028] Figure 1 A schematic diagram of the structure of the multi-party intelligent supervision system for residual mud and slag provided by the present invention. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] The embodiment of the present invention discloses a multi-party intelligent supervision system for residual mud and slag, which mainly constructs a multi-party collaborative supervision platform for realizing data sharing and cooperation among different departments, enterprises and the public, improving information transparency; and reducing manual monitoring by managers, turning it into an assessment of the construction site, so that supervisors can access data and receive early warning information anytime and anywhere, and improve the flexibility and timeliness of decision-making. The present invention encourages the public to participate in earthwork management and supervision, feedback problems in a timely manner, and enhance the public's supervision awareness through the disclosure of the tracks of residual mud and slag vehicles; and it can dispatch residual mud resources and dispatch residual mud to areas lacking mud. Residual mud and slag can be recycled to reduce the exploitation of natural resources.

[0031] In one embodiment, a multi-party intelligent supervision system for residual mud and slag is provided. Figure 1 , including a data acquisition module, a data processing module and a multi-party supervision platform. The data processing module is used to obtain the mud output, transportation times, transportation routes, leakage conditions and soil scheduling quantity according to the collected data; and send the mud output, transportation times, transportation routes, leakage conditions and soil scheduling quantity to the multi-party supervision platform.

[0032] In this embodiment, the data acquisition module includes a planning data acquisition unit, a construction site data acquisition unit, a market transaction data acquisition unit, a traffic video data acquisition unit, a multi-scene sensor data acquisition unit, and an environmental weather data acquisition unit.

[0033] Among them, the planning data collection unit is used to collect terrain and geological data, construction design data, environmental protection and regulatory data, surrounding planning data, etc. Through the analysis and prediction of the data, the residual mud and soil can be reasonably estimated and managed. The data source is the Natural Planning Bureau.

[0034] The construction site data collection unit is used to collect excavation and filling data, including excavation depth, area and volume, which directly affect the amount of residual mud and slag generated; soil and geological data, which describe soil type, density, water content, etc., and determine the treatment method and transportation requirements of slag; construction progress and stage data, which reflect the earthwork arrangement at different construction stages and affect the time and transportation plan of slag generation; transportation and processing data: including transportation routes, vehicle capacity and disposal methods of slag, to ensure that residual mud and slag are reasonably handled and distributed. The data source is the construction site implementation unit.

[0035] The market transaction data collection unit is used to collect the cost data of slag treatment: including the market prices of transportation, treatment and landfill, which affect the total cost budget of the project; resource recovery and reuse data: the market demand and price of reusable materials in slag (such as crushed stone and concrete), which affect the decision of resource recycling; supply and demand relationship data: the demand of different projects for residual mud and slag, such as the demand for slag in filling projects and the mud output of supply projects, which determines the scale of transactions; policy and regulatory data: government subsidies, tax policies and treatment standards related to slag transactions in the market, which affect the compliance and economy of the transaction process. The data source is the open market transaction information website.

[0036] The traffic video data acquisition unit is used to collect information including road traffic flow, traffic congestion, public transportation routes, etc.; it contains road traffic monitoring data, GPS trajectory data, etc. to help identify busy traffic areas and traffic bottlenecks.

[0037] In addition to the above traffic data, traffic checkpoint video data is also required, including vehicle monitoring data: video monitoring is used to record the entry and exit of muck transport vehicles to ensure that the vehicles are operated at legal times and locations. Load monitoring data: video analysis can assist in confirming the vehicle's load, prevent overloaded transportation, and ensure compliance with traffic regulations. Transport route tracking data: monitor the route of transport vehicles to ensure that they follow the prescribed route to avoid illegal dumping or illegal dumping of muck. Safety compliance data: record safe operations during transportation to ensure that drivers comply with safety regulations, reduce accident risks, and ensure legal operations. The data source is the traffic management department.

[0038] The multi-scenario sensor data acquisition unit is used to collect data generated by earthwork at the construction site: the earthwork sensors at the construction site monitor the amount of earthwork excavated and filled in real time, and record the volume of residual mud and slag generated; construction site exit measurement data: monitor the load, loading volume and loading conditions of the exit transport vehicles to ensure the accurate transportation of residual mud and slag; disposal site entry measurement data: record the amount of slag received by the disposal site, entry time and processing capacity, to ensure the legal and effective disposal of slag; construction site and disposal site environmental monitoring data: monitor the environmental parameters of the construction site and disposal site (such as dust, noise, etc.) to ensure that the residual mud and slag treatment process meets environmental protection standards; transportation vehicle tracking data: monitor the operation path, time and arrival of slag transport vehicles between the construction site and the disposal site through GPS and other sensors to ensure the safe and efficient transportation process. The data comes from the urban management and housing and construction departments.

[0039] The environmental weather data collection unit is used to collect precipitation data: monitor rainfall, affect soil moisture and earthwork excavation, transportation and slag treatment, and avoid mud loss caused by rain. Temperature and humidity data: affect the dry and wet state of the soil, and then affect the difficulty of excavation, transportation efficiency and treatment of slag. Wind speed and direction data: Meteorological warning data: provide warnings of extreme weather (such as heavy rain, typhoons, etc.), affect construction arrangements and slag treatment plans, and reduce environmental risks. The data source is the meteorological department.

[0040] In order to further optimize the above technical solution, feature extraction is performed on the acquired data to facilitate subsequent data processing and improve the performance and generalization ability of the entire regulatory system; specifically,

[0041] For planning data features, elevation, slope, aspect, surface cover type, etc. are extracted from terrain data. These features can help understand the difficulty of excavation and filling of earthwork. Soil type, soil density, groundwater level, etc. are extracted from geological data. These features affect the characteristics and treatment methods of the waste soil. Excavation depth, area, design drawings, etc. are extracted from construction design data to predict the volume of excavation and filling. Compliance requirements, environmental impact assessments, etc. from environmental protection and regulatory data determine the legality and method of waste soil treatment. Neighboring facilities, traffic conditions, land use, etc. are obtained from surrounding planning data to affect transportation and treatment plans.

[0042] Site data features, obtained from excavation and filling data, include excavation depth, filling area and volume. These data directly affect the amount of residual mud and debris generated. Soil and geological data are obtained to describe soil type, density, water content, etc. These characteristics determine the characteristics and treatment requirements of the debris. The arrangements reflecting different construction stages are obtained from construction progress and stage data, which helps to understand the earthwork work plan and its impact on residual mud and debris. The transportation and treatment data involve transportation routes, vehicle capacity and disposal methods of debris, which affect the reasonable treatment and distribution of debris.

[0043] Market transaction data characteristics: extract transportation costs from cost characteristics, extract average costs for different transportation distances, and analyze the impact of transportation routes on costs. Extract treatment costs: extract average costs of different treatment methods (such as landfill and reuse) from treatment price data. Extract landfill costs: obtain the market price of landfills as an important part of project costs. Extract reusable material prices from resource recovery characteristics, extract market prices of materials such as crushed stone and concrete, and analyze the recycling value of resources. Extract market demand data, obtain the market demand for recycled materials, and evaluate the market potential of resources. Extract demand characteristics from supply and demand relationship characteristics, extract specific demand for slag in landfill projects to reflect changes in demand for different projects. Extract supply characteristics, record the mud output of different projects, and analyze the relationship between supply and demand. Extract transaction scale characteristics, calculate the total transaction volume within a specific time, and analyze the market activity. Extract subsidy characteristics from policy and regulatory characteristics: extract relevant government subsidy policies and evaluate their economic impact on slag transactions. Extract tax policy characteristics, obtain tax policy information applicable to slag transactions, and analyze the impact of taxes on costs and pricing. Extract processing standard features and regulatory standards for different processing methods to ensure transaction compliance.

[0044] Traffic video data features: extract the number of vehicles passing through a certain location per unit time from traffic counts to analyze the changing trend of traffic flow. Extract peak hour identification: identify the peak hours of traffic flow based on traffic data to help traffic management. Extract congestion duration from traffic congestion features, calculate the duration of traffic congestion in a specific area, and evaluate traffic efficiency. Extract congestion frequency: analyze the frequency of congestion at a certain location and identify traffic bottlenecks. Extract entry and exit records from vehicle monitoring features: record the entry and exit time and frequency of muck transport vehicles through video monitoring to ensure that vehicles operate within the specified time. Extract license plate recognition: extract vehicle license plate information for tracking and management of specific vehicles. Extract load data from load monitoring features: obtain the vehicle's load through video analysis to determine whether it is overloaded. Extract compliance checks: evaluate whether the vehicle's load complies with relevant traffic regulations. Extract driving routes from transportation path features: extract vehicle GPS trajectory data, and analyze the deviation between its actual driving route and the specified route. Extract route optimization analysis: calculate the time and distance of the transportation path to optimize the efficiency of muck transportation. Extract safety operation records from safety compliance features to record safety operations during transportation, such as traffic light compliance, emergency braking, and speeding. Extract driver behavior analysis to analyze the driver's driving behavior during transportation and assess safety risks.

[0045] Multi-scenario sensor data features: extract the excavation and filling volume from the earthwork generation features, record the excavation and filling data of the construction site in real time, and extract the total earthwork volume on a daily, weekly or monthly basis. Extract the volume of residual mud and slag, calculate the volume of residual mud and slag generated, and track its changing trend to evaluate the progress of the project. Extract load monitoring from the exit measurement features, extract the load information of the transport vehicle, and ensure that the loading volume of each exit is within the compliance range. Extract loading situation analysis to analyze whether there is insufficient or overload during the loading process to ensure the accuracy of transportation. Extract the amount of slag received from the disposal site entry measurement features, record the amount of slag received by the disposal site, and extract daily and weekly entry data. Extract the entry time features, extract the time information of the slag entry, and analyze the peak and trough periods of slag treatment. Extract the processing capacity, record the processing capacity of the disposal site, and analyze its processing efficiency for the incoming slag. Extract dust concentration from the environmental monitoring features, extract dust concentration data of the construction site and the disposal site, and evaluate the environmental impact. Extract noise level and monitor the noise level to ensure that the treatment process meets environmental protection standards. Extract environmental compliance index, construct environmental compliance indicators, and comprehensively evaluate the compliance of various environmental data. Extract GPS path data from the tracking characteristics of transport vehicles, extract the GPS track of transport vehicles, and record the actual driving path. Extract transportation time, calculate the transportation time from the construction site to the disposal site, and evaluate transportation efficiency. Extract arrival status, monitor the arrival time of transport vehicles, and ensure on-time arrival and unloading of soil.

[0046] Environmental weather characteristics: extract the total amount of precipitation from the precipitation characteristics, extract the total precipitation in a specified time period, and use it to evaluate the impact on soil moisture. Extract precipitation intensity, analyze the intensity of precipitation (such as mm / hour), and affect the risk assessment of mud loss. Extract precipitation duration, record the duration of precipitation events, and analyze its impact on construction progress. Extract daily average temperature and humidity from temperature and humidity characteristics, extract daily average temperature and humidity data, and analyze its impact on soil status. Extract soil moisture, combine temperature and humidity data to estimate the dry and wet state of the soil to determine the difficulty of excavation. Extract temperature change rate, calculate the temperature change rate, and evaluate the short-term impact on soil properties. Extract average wind speed from wind speed and wind direction characteristics, and extract average wind speed in a specific time period to evaluate construction safety. Extract wind direction changes, record wind direction changes, and analyze their impact on the construction environment (such as dust). Extract warning event records from meteorological warning characteristics, and record the occurrence time and type of extreme weather events (such as heavy rain and typhoons). Extract warning impact analysis, and analyze the impact of meteorological warnings on construction and slag treatment plans based on historical data.

[0047] In one embodiment, the multi-party supervision platform includes a market supervision platform, a government supervision platform, and a public supervision platform. This application sends the mud output, transportation times, transportation routes, and leakage to the government supervision platform; sends the transportation route to the public supervision platform; sends the transportation route and soil scheduling volume to the market supervision platform; so that the multi-party supervision system can realize the whole process monitoring of the residual mud and slag, improve the supervision efficiency, protect the environment, and promote sustainable development.

[0048] Specifically, the government needs to accurately grasp the amount of mud discharged from the construction site in order to formulate a reasonable slag treatment plan and resource scheduling strategy. By calculating the amount of mud discharged from the construction site, the government can effectively supervise the construction site and ensure that resources are used reasonably.

[0049] At the same time, the government needs to know the number of times the muck is transported so that it can effectively monitor and manage the muck trucks. By calculating the number of times the muck is transported, the government can ensure that the muck trucks transport the muck at the prescribed number of times and avoid illegal behaviors such as overloading and over-limit.

[0050] In addition, route planning is a key link in multi-party supervision. The government needs to plan the transportation routes of muck trucks based on real-time traffic data and geographic information to ensure the safety and efficiency of transportation.

[0051] The government also needs to monitor the vibration amplitude and mud weight of the vehicle to determine whether the muck truck will leak mud. This step helps the government to detect and deal with the leakage problem in time and protect the environment from pollution.

[0052] For the market side, it is necessary to optimize the logistics route according to route planning and reduce transportation costs; and it is necessary to dispatch and track the soil usage according to the soil usage in different regions, market transaction volume and mudwork volume of the construction site. This step helps the market side to reasonably allocate resources, improve resource utilization efficiency and promote sustainable development.

[0053] For public supervision, the transparency of supervision can be improved by obtaining certain route information and understanding the transportation situation of dump trucks.

[0054] In one embodiment,

[0055] 1. Calculation of mud output:

[0056] Calculating the amount of mud produced at a construction site usually involves multiple factors, such as the area and depth of the excavation area and the density of the soil. The following are the general steps for calculating the amount of mud produced at a construction site

[0057] 1.1 Determine the volume of the excavation area

[0058] 1.2 Consider soil expansion coefficient (loose volume coefficient)

[0059] The soil will usually expand after excavation, so the soil bulk coefficient needs to be considered. The bulk coefficient is usually 1.2 to 1.3, which means that the volume of the excavated soil will increase by 20% to 30% compared to the original soil volume;

[0060] Mud discharge volume VactualV_{\text{actual}}Vactual:

[0061] Vactual=V×Loose square coefficientV_{\text{actual}}=V\times\text{Loose square coefficient}Vactual=V×Loose square coefficient.

[0062] 1.3 Consider soil type

[0063] Soil density and structure (such as sand, clay, gravel, etc.) will also affect the calculation of mud output, and some soil types have higher expansion coefficients.

[0064] 2. Calculation formula for the number of muck transportation times:

[0065] The calculation of the number of muck transportation times usually depends on the amount of muck generated, the capacity of the transport vehicle, and the efficiency of transportation. The steps to calculate the number of muck transportation times are:

[0066] 2.1 Determine the total amount of slag

[0067] The total amount of slag is estimated based on the total amount of all the residual mud and slag generated at the construction site or in a certain period of time. The unit is usually cubic meters (m 3 ) or tons.

[0068] 2.2 Determine the capacity of each transport vehicle

[0069] The capacity of each transport vehicle is usually fixed and is measured in cubic meters (m 3 ) or tons. If the density of the soil is known, conversion between different units can be performed.

[0070] 2.3. Calculation of theoretical transport times

[0071] Use the formula to calculate the theoretical number of muck transportation:

[0072] Number of transports = total amount of soil (m 3 or tons) capacity of each vehicle (m 3 or tons)\text{Number of transports}=\frac{\text{Total amount of slag (m 3 or tons)}}{\text{The capacity of each vehicle (m 3 or tons)}}Transport times = capacity of each vehicle (m 3 or tons) Total amount of slag (m 3 or tons)

[0073] 2.4 Consider transportation efficiency

[0074] Actual shipping times may be affected by a number of factors:

[0075] Road conditions: If the road is congested or in poor condition, it may affect the efficiency of transportation.

[0076] Number of vehicles: If there are multiple vehicles working at the same time, the actual transportation frequency will be higher.

[0077] Shipping routes: Short-distance shipments allow for quick turnaround, while long-distance shipments increase the time required for each shipment.

[0078] 2.5 Adjustment of actual transport times

[0079] In addition to the theoretical number of transports, the actual situation of transport must also be considered, such as the number of vehicle turnovers in a day. Assuming that each vehicle transports multiple times a day, the total daily transport volume can be calculated:

[0080] Daily transportation volume (times) = total amount of slag / single transportation capacity × number of vehicles × daily transportation frequency

[0081] 3. Path planning formula:

[0082] The route planning for residual mud and soil transportation involves many factors, including the location of the starting and ending points, traffic conditions, road conditions, vehicle types and capacity, etc. The fastest route and method are selected through the following calculations:

[0083] 3.1 Shortest Path Problem

[0084] The shortest path problem is usually solved using algorithms from graph theory, such as Dijkstra's algorithm and A* algorithm.

[0085] Dijkstra algorithm: Applicable to the case where there are non-negative edge weights in the graph.

[0086]

[0087] A algorithm*: speeds up search through heuristic estimation, suitable for complex scenarios.

[0088] f(n)=g(n)+h(n)

[0089] Among them, g(n) is the actual cost from the starting point to the current node, and h(n) is the heuristic estimated cost from the current node to the target node.

[0090] 3.2 Calculation of transportation costs

[0091] The choice of transportation route usually takes into account the transportation cost, which can be calculated using the following formula:

[0092] Transportation cost = transportation distance × cost per kilometer × number of vehicles

[0093] 3.3 Consider traffic flow and road conditions

[0094] When planning a route, you can introduce traffic flow and road conditions to modify the cost of the route:

[0095] Adjusted transportation cost = transportation distance × (cost per kilometer + traffic impact coefficient)

[0096] 3.4 Dynamic Path Planning

[0097] In some cases, real-time traffic data may affect the selection of the best path, which requires a dynamic path planning algorithm (such as a dynamic programming algorithm) to adjust the path in real time.

[0098] 3.5 Constraints

[0099] There may be some constraints during transportation, such as:

[0100] Vehicle weight limit: You need to ensure that each vehicle does not exceed its weight limit.

[0101] Construction areas and restricted areas: Avoid construction areas and restricted areas.

[0102] 3.6 Comprehensive Model

[0103] Linear programming (LP) or integer programming (IP) models can be used to optimize path selection by taking multiple factors into consideration. For example:

[0104] minimize

[0105] Among them C ij is the transportation cost from node i to node j, X ij is the decision variable (0 or 1) for whether to choose this path.

[0106] 4. Determine whether there is leakage:

[0107] Determining whether a sludge truck is leaking usually involves monitoring and analyzing multiple factors during the transportation process.

[0108] Here are some commonly used methods and formulas:

[0109] 4.1 Weight change method formula:

[0110] ΔW=W 装载 -W 卸载

[0111] Explanation: If the unloaded weight is significantly lower than the loaded weight, this may indicate spillage. You can set a threshold to determine whether the difference is significant.

[0112] 4.2. Calculation formula for vehicle load factor:

[0113]

[0114] Explanation: If the load factor is less than expected, it may mean that some of the debris has leaked.

[0115] 5. Soil usage scheduling formula: Assume there are n areas, and the soil usage of each area is U i , the amount of mudwork on the construction site is D, then the scheduling amount for each area is T i =U i -D / n.

[0116] In one embodiment, the execution code of the data processing module is:

[0117]

[0118]

[0119]

[0120] In order to further optimize the above technical solution, this application constructs a neural network model to assist the government supervision platform in supervision. The neural network model is used to predict the amount of slag generated in each area or the amount generated by the slag truck based on the residual mud and slag distribution map and the residual mud and slag truck bucket, and to predict the future generation based on historical earthwork volume and weather data.

[0121] In this embodiment, the neural network model includes a convolutional neural network (CNN) and a recurrent neural network (RNN);

[0122] In an exemplary embodiment,

[0123] (I) Building a CNN model using the Keras deep learning framework

[0124] Through this CNN model, the system can predict the amount of sludge generated in each area and the amount of sludge in each vehicle based on the input residual mud and sludge distribution map and residual mud and sludge truck bucket, helping the government to conduct multi-party supervision. You can adjust the number of convolutional layers, pooling layers, and fully connected layers or the number of neurons according to the characteristics of the specific data set:

[0125] 1. Input layer: Design an appropriate input layer to process the input data. For image data, it is usually a three-dimensional shape (width, height, number of channels), and for time series data, it may be a two-dimensional or three-dimensional shape (time step, feature dimension, number of channels). The input layer accepts feature data related to residual mud and slag, which may include vehicle monitoring, license plate recognition, vehicle size, trajectory, etc. These feature data may be two-dimensional or three-dimensional.

[0126] 2. Convolutional layer and pooling layer: to improve the representation ability of the model. Add a Flatten layer before the fully connected layer to flatten the feature map. The convolutional layer is used to extract feature information. The input data is convolved by a sliding convolution kernel to generate a feature map. The pooling layer is used to reduce the dimension of the feature map, reduce the amount of calculation and retain important information. Usually, the maximum pooling or average pooling operation is used.

[0127] 3. Fully connected layer for classification or regression tasks. The output of the pooling layer is flattened into a one-dimensional vector as the input of the fully connected layer. The fully connected layer is used to learn the relationship between features. It can contain multiple hidden layers and neurons, and introduce nonlinearity through activation functions.

[0128] 4. Output layer: The output layer is determined according to the task requirements. It can be one node for binary classification tasks (such as whether the size of the residual mud and slag vehicle is appropriate), or multiple nodes for multi-classification tasks (such as the trajectory of the residual mud and slag vehicle). For binary classification tasks, the sigmoid activation function is usually used; for multi-classification tasks, the softmax activation function is usually used.

[0129] The specific operations are as follows:

[0130] Assume that there is image data of a sludge truck (each image represents a heat map of sludge truck generation), the size of each image is 64×6464\times 6464×64 pixels, and the image has one channel (grayscale image). The goal is to predict the amount of sludge truck generation to determine whether the sludge truck is overweight.

[0131] The number of nodes in the input layer should match the dimension of the features. In CNN, this is usually expressed as the width, height, and number of channels of the image. Below is an example CNN model showing how to build a model by stacking convolutional layers, pooling layers, and fully connected layers.

[0132] The code is as follows:

[0133]

[0134]

[0135] You need to prepare training and test data. Suppose you have the following dataset:

[0136] X_train: heat map of residual mud and slag in the training set, with the shape of (number of samples, 64, 64, 1)

[0137] y_train: the corresponding amount of residual mud and soil generated, the shape is (number of samples,)

[0138] X_test: heat map of residual mud and slag in the test set, with the shape of (number of samples, 64, 64, 1)

[0139] y_test: the corresponding amount of residual mud and soil generated, the shape is (number of samples,)

[0140] Training the model

[0141] model.fit(X_train,y_train,epochs=10,batch_size=32,

[0142] validation_split = 0.2)

[0143] Evaluating the Model

[0144] test_loss,test_mae=model.evaluate(X_test,y_test)

[0145] print(f"Test MAE:{test_mae}")

[0146] (II) Building RNN Model Using Keras Deep Learning Framework

[0147] Recurrent Neural Networks (RNNs) are used to process sequence data. Their key feature is that they can retain state information from past time steps, thereby capturing temporal dependencies in input data. Since earthwork generation, transportation volume, weather data, etc. may be time-dependent in the residual mud and slag supervision system, RNNs are particularly suitable for these sequence prediction tasks. For example: predicting earthwork generation based on historical earthwork volume, weather conditions, and other data. Predicting future transportation volume and efficiency based on transportation routes and transportation history to predict transportation volume.

[0148] 1. Input layer: Receives time series data, usually in the form of a two-dimensional array (number of samples × time step × feature dimension). In the residual mud and slag system, the input data may include daily earthwork volume, weather data, transportation history and other features.

[0149] 2. Hidden layer: RNN retains information from past time steps through recursive connections. Traditional RNNs are prone to the vanishing gradient problem, so LSTM (Long Short-Term Memory Network) or GRU (Gated Recurrent Unit) are usually used to replace the traditional RNN layer. LSTM and GRU can effectively capture long-term dependency features by introducing memory gates and forgetting mechanisms. The LSTM structure is mainly used, including input gates, forget gates, and output gates, to solve the long-term dependency problem by controlling the flow of information.

[0150] 3. Fully connected layer: The output of the hidden layer is flattened into a one-dimensional vector and processed by the fully connected layer. The fully connected layer captures the nonlinear relationship between features through the activation function and is suitable for classification or regression tasks.

[0151] 4. Output layer: Select the appropriate activation function according to the task requirements: Regression tasks (such as predicting future earthwork volume): linear activation functions can be used to output continuous values. Classification tasks (such as whether there is illegal transportation): sigmoid (binary classification) or softmax (multi-classification) can be used.

[0152] The specific operations are as follows:

[0153] Prepare time series data. Assume that the data includes historical earthwork production, weather data, transportation records, etc., and the shape is (number of samples, time step, feature dimension).

[0154] X_train: training data, shape is (number of samples, time steps, feature dimensions).

[0155] For example, each sample contains 10 days of earthwork production and other relevant features.

[0156] y_train: target data, of shape (number of samples,), for example, predicting the amount of earthwork generated on a certain day in the future.

[0157] X_test and y_test: test data, with a similar structure to the training data.

[0158] Create an LSTM model and specify the input shape, number of layers, activation function, etc.

[0159] The code is as follows:

[0160] import tensorflow as tf

[0161] from tensorflow.keras.models import Sequential

[0162] from tensorflow.keras.layers import LSTM,Dense

[0163] #Create RNN model

[0164] model=Sequential()

[0165] #LSTM layer (assuming each sample contains 10 days of data, 5 features per day)

[0166] model.add(LSTM(units=50,return_sequences=False,input_shape=(10, 5)))

[0168] #Fully connected layer

[0169] model.add(Dense(units=64,activation='relu'))

[0170] #Output layer: regression task, output earthwork generation

[0171] model.add(Dense(units=1,activation='linear'))

[0172] # Compile the model

[0173] model.compile(optimizer='adam',loss='mean_squared_error',

[0174] metrics = ['mae'])

[0175] Train the model, use historical data to train the model, and gradually optimize the model performance by setting the batch size and the number of training iterations.

[0176] #Assume that X_train is training data with shape (number of samples, time steps, feature dimensions)

[0177] #y_train is the corresponding earthwork volume

[0178] model.fit(X_train,y_train,epochs=20,batch_size=32,

[0179] validation_split = 0.2)

[0180] Epochs: The number of training cycles, which can be adjusted according to the size of the dataset and the complexity of the model.

[0181] Batch_size: The number of samples per training.

[0182] validation_split: Split a portion of the training data for validation of the model.

[0183] Evaluate the model and evaluate the performance of the model on the test data, calculating the mean squared error (MSE) and mean absolute error (MAE).

[0184] test_loss,test_mae=model.evaluate(X_test,y_test)

[0185] print(f"Test MAE:{test_mae}")

[0186] Predict future earthwork generation. Once the model is trained, use new data to make predictions to help regulators optimize earthwork management plans.

[0187] predictions=model.predict(new_data)

[0188] new_data is the input data that the model receives from future time steps, and it should have the same dimensional structure as the training data.

[0189] Adjust the model

[0190] You can adjust the number of LSTM units and increase or decrease the number of model layers to improve the model's fitting ability.

[0191] Add Dropout layer to prevent the model from overfitting.

[0192] Changing the optimizer (e.g. using RMSprop, Adam, etc.) to improve the learning speed and accuracy of the model.

[0193] Preferably, compile the model using an appropriate loss function and optimizer. For binary classification tasks, you can use the binary_crossentropy loss function; for multi-classification tasks, you can use the categorical_crossentropy loss function; for regression tasks, you can use the mse loss function.

[0194] 1. Choose an optimizer: Choose an optimizer that is suitable for the model and data to adjust the model weights to minimize the loss function. The Adam optimizer is a common choice that can usually effectively adjust the learning rate and adapt to different data sets and model structures.

[0195] 2. Select the loss function: Select the appropriate loss function according to the task type of the model. For regression tasks, the mean_squared_error loss function can be used because the goal is to predict the amount of residual mud and soil, which is a regression task. You can use the mean square error (MSE) as the loss function and select a suitable optimizer.

[0196] 3. Select evaluation metrics: Select one or more metrics to evaluate the performance of the model. For classification problems, common evaluation metrics include accuracy, precision, recall, and F1 score.

[0197] 4. Compile Model: Compile the model using the optimizer, loss function, and evaluation metric selected above. model.compile(optimizer = optimizer, loss = loss, metrics = metrics)

[0198] The model analysis conducted by the present invention based on the overall data can help improve the efficiency of the construction site, reduce environmental pollution, optimize the processing of soil and slag, and improve the level of construction site management, thereby providing a better decision-making basis for construction site managers. The model has a wide range of application prospects, specifically:

[0199] 1) The purpose of this study is to enable urban management personnel to monitor the generation, transportation and disposal of residual mud and soil in real time through technical means, and to reduce environmental pollution caused by improper treatment of residual mud and soil through reasonable supervision measures.

[0200] 2) After the model is built and trained, it will be embedded in urban planning, transportation, urban management, housing and construction data according to the city's development situation. It is necessary to pre-enter the historical data of the city's residual mud and slag, and coordinate the corresponding planning data, construction site data, market transaction data, traffic video data, multi-scene sensor data, environmental weather data, etc., to establish a multi-party intelligent supervision system for residual mud and slag, including residual mud and slag transaction management, residual mud and slag monitoring management, and residual mud and slag vehicle route management.

[0201] 3) The management of residual mud and slag transactions aims to improve the management efficiency, transparency and recycling of residual mud and slag. The system includes modules such as market information release, transaction process management, quality monitoring, transportation scheduling and user services. Through market information release, users can obtain supply and demand information in real time and optimize decisions. Transaction process management simplifies the quotation and settlement links and improves transaction efficiency. Quality monitoring ensures that the traded earth meets environmental protection standards and reduces environmental risks. The transportation scheduling module optimizes logistics routes and reduces transportation costs. User services provide technical support and consulting to enhance user experience. The system integrates data analysis to help users predict market trends and make more reasonable decisions. The overall effect is to achieve efficient use of resources, promote sustainable development, and enhance the government's regulatory capabilities. Ultimately, the system provides technical support for the standardization and scientific management of the residual mud and slag industry.

[0202] 4) Monitoring and management of residual mud and slag is to enhance the government's ability to supervise the disposal and transportation of residual mud and slag. The system includes modules such as real-time monitoring, data analysis, report generation and regulatory compliance. Through real-time monitoring, government managers can obtain information on the generation, transportation and treatment of residual mud and slag in a timely manner to ensure compliance in all links. The data analysis function helps managers identify abnormal situations, formulate targeted measures, and improve the scientific nature of management decisions. The report generation module regularly provides detailed supervision reports for easy review and evaluation. The regulatory compliance module ensures that all operations comply with environmental regulations and reduce legal risks. The system improves the transparency of information through a visual interface, making it easier for government managers to understand and use. Ultimately, the system improves the efficiency and effectiveness of the government's management of residual mud and slag, and helps promote sustainable development. Through scientific supervision, the government can better respond to environmental challenges and protect ecological security. This system also enhances the public's trust in government supervision and promotes social harmony.

[0203] 5) The route management of residual mud and slag vehicles is to enhance transparency and public trust, and ensure the compliance and safety of residual mud and slag transportation. The system content includes functions such as real-time route monitoring, public transportation record disclosure, abnormal behavior alarm and statistical analysis. Through real-time route monitoring, government managers can grasp the actual transportation route of vehicles to ensure that they comply with regulations. The public transportation record enables the public to query vehicle transportation information and enhance the transparency of supervision. The abnormal behavior alarm function can identify non-compliant behaviors in a timely manner and respond quickly to potential problems. The statistical analysis module helps managers evaluate transportation efficiency and environmental compliance, and provides data support for decision-making. The implementation of this system effectively reduces violations in the process of slag transportation and improves the environmental protection effect. The information disclosed to the outside also enhances the environmental awareness of citizens and promotes social participation. By establishing a good information communication mechanism, the government can carry out residual mud and slag management more effectively. Ultimately, the system helps to build a harmonious urban environment and achieve sustainable development goals.

[0204] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0205] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-party intelligent supervision system for residual mud and slag, characterized in that: Including data collection module, data processing module and multi-party supervision platform, Data collection module, used to collect residual mud and slag transportation data; The data processing module is used to obtain the mud output, transportation times, transportation routes, leakage conditions and soil scheduling amount based on the collected data; The multi-party supervision platform is used to receive and supervise the residual mud and soil based on the mud discharge volume, transportation times, transportation routes, leakage conditions and soil scheduling volume.

2. The multi-party intelligent supervision system for residual mud and slag according to claim 1 is characterized in that: The data acquisition module includes a planning data acquisition unit, a construction site data acquisition unit, a market transaction data acquisition unit, a traffic video data acquisition unit, a multi-scene sensor data acquisition unit and an environmental weather data acquisition unit.

3. The multi-party intelligent supervision system for residual mud and slag according to claim 2 is characterized in that: The planning data collection unit is used to collect topographic and geological data, construction design data, environmental protection and regulatory data, and surrounding planning data; The site data collection unit is used to collect excavation and filling data, soil and geological data, construction progress and stage data, transportation and processing data; The market transaction data collection unit is used to collect slag treatment cost data, resource recovery and reuse data, supply and demand relationship data, policy and regulatory data; The traffic video data acquisition unit is used to collect road traffic flow, traffic congestion, public transportation routes, traffic checkpoint video data, load monitoring data, transportation route tracking data and safety compliance data; The multi-scenario sensor data acquisition unit is used to collect construction site earthwork data, construction site exit measurement data, disposal site entry measurement data, construction site and disposal site environmental monitoring data, and transport vehicle tracking data; The environmental weather data collection unit is used to collect precipitation data, temperature and humidity data, wind speed and wind direction data.

4. The multi-party intelligent supervision system for residual mud and slag according to claim 1 is characterized in that: The multi-party supervision platform includes market supervision platform, government supervision platform and public supervision platform.

5. The multi-party intelligent supervision system for residual mud and slag according to claim 4 is characterized in that: The government supervision platform supervises based on mud discharge, number of transportation times, transportation routes, and leakage; the public supervision platform supervises based on the transportation routes; and the market supervision platform supervises based on the transportation routes and soil scheduling quantities.

6. The multi-party intelligent supervision system for residual mud and slag according to claim 1 is characterized in that: A neural network model is constructed to assist the government supervision platform in supervision. The neural network model is used to predict the amount of slag generated in each area or the amount generated by the slag truck based on the residual mud and slag distribution map and the residual mud and slag truck bucket, and to predict the future generation based on the historical earthwork volume and weather data.

Images