Commercial concrete shared wagon balance site selection method, device and equipment and storage medium

Abstract

The invention relates to a commercial concrete shared wagon balance site selection method, device and equipment and a storage medium, and belongs to the technical field of facility site selection, and the commercial concrete shared wagon balance site selection method comprises the steps: obtaining commercial concrete demand data and wagon balance state data of a commercial concrete distribution area, drawing a commercial concrete demand diagram and a wagon balance state diagram of the commercial concrete distribution area based on the commercial concrete demand data and the wagon balance state data; a preset double-branch neural network model is adopted to recognize the commercial concrete demand diagram and the wagon balance state diagram, an action value diagram is obtained, and the action value diagram is used for reflecting the action value corresponding to the site selection action of the commercial concrete shared wagon balance in the commercial concrete distribution area; and carrying out commercial concrete shared wagon balance site selection based on the commercial concrete shared wagon balance site selection action with the highest action value in the action value map. According to the method, complex commercial concrete site selection factors can be considered, changing commercial concrete requirements can be responded in real time, an efficient site selection strategy is provided, energy consumption and unreasonable consumption in commercial concrete transportation are reduced, and cost is saved.

Description

Technical Field

[0001] The present invention relates to the technical field of facility site selection, and in particular to a method, device, equipment and storage medium for site selection of a commercial concrete shared weighbridge. Background Art

[0002] With the development of my country's construction industry, the commercial concrete (MCC) supply industry has become increasingly mature. In order to ensure that the MCC is supplied in sufficient quantity according to the order, the orderer of MCC needs to weigh the concrete transport vehicle, and the scale is the main tool for weighing the concrete transport vehicle. The traditional solution is a "one-to-one" type, that is, a separate scale is set up on each construction site, and a scale only serves one construction site. The advantage of the traditional solution is that since it is put into use immediately after weighing, it can accurately detect the unreasonable large losses in the transportation of MCC; however, the disadvantage of the traditional solution is that setting up a separate scale for each construction site will bring higher costs. Some construction sites are close to each other. Within a short distance, there is no need to worry about the transportation of MCC causing large and unreasonable losses. Therefore, it is reasonable and feasible to use the same scale to weigh the MCC transport vehicles at construction sites close to each other. Compared with the traditional "one-to-one" solution, it can effectively reduce costs.

[0003] However, the site selection for this shared scale has become a new problem, because different construction sites have different demands for ready-mixed concrete, the ready-mixed concrete supply capabilities of ready-mixed concrete plants are also different, and the distance between the ready-mixed concrete plant, construction site, and shared scale, as well as the difficulty of transportation, are all factors to be considered in the site selection of shared scales. In addition, the demand for ready-mixed concrete is constantly changing, so the location of the shared scale may need to be changed frequently. However, the cost of frequently changing the shared scale may be higher than that of building a new scale. Therefore, how to coordinate the site selection of a shared scale is a difficult point.

[0004] It can be seen that the existing technology cannot provide an accurate site selection plan for commercial concrete shared weighbridges. Summary of the invention

[0005] In view of this, it is necessary to provide a method, device, equipment and storage medium for site selection of a commercial concrete shared weighing scale to solve the problem that the existing technology cannot provide an accurate site selection plan for a commercial concrete shared weighing scale.

[0006] In order to solve the above problems, the present invention provides a method for selecting a site for a commercial concrete shared weighbridge, comprising: Obtain the commercial concrete demand data and weighbridge status data of the commercial concrete distribution area, and draw the commercial concrete demand graph and weighbridge status graph of the commercial concrete distribution area based on the commercial concrete demand data and weighbridge status data; The preset dual-branch neural network model is used to identify the commercial concrete demand graph and the weighbridge status graph to obtain an action value graph, which is used to reflect the action value corresponding to the commercial concrete shared weighbridge site selection action in the commercial concrete distribution area; The site selection for the commercial concrete shared weighing scale is performed based on the commercial concrete shared weighing scale site selection action with the highest action value in the action value graph.

[0007] In a possible implementation, the commercial concrete demand data and the weighbridge status data of the commercial concrete distribution area are obtained, and the commercial concrete demand graph and the weighbridge status graph of the commercial concrete distribution area are drawn based on the commercial concrete demand data and the weighbridge status data, including: Grid the commercial concrete distribution area, and map the commercial concrete plants, commercial concrete scales and construction sites in the commercial concrete distribution area into the grid area; Determine the waybill data of each construction site within a preset time step, and determine the pixel value of the grid corresponding to the construction site based on the waybill data to obtain a commercial concrete demand map; The active state of each commercial concrete scale is determined, and the pixel value of the corresponding grid of the commercial concrete scale is determined based on the active state to obtain a scale state diagram.

[0008] In a possible implementation, a preset dual-branch neural network model is used to identify the commercial concrete demand graph and the weighbridge status graph to obtain an action value graph, including: The first branch network of the preset double-branch neural network model is used to extract features of the commercial concrete demand graph to obtain a commercial concrete demand feature graph; The second branch network of the preset double-branch neural network model is used to extract features from the weighbridge state diagram to obtain a weighbridge state feature diagram; The feature fusion graph of commercial concrete demand feature graph and the feature graph of weighbridge status are performed to obtain a feature fusion graph, and the action value graph is determined based on the feature fusion graph.

[0009] In a possible implementation, determining an action value graph based on a feature fusion graph includes: Based on the feature fusion graph, a reinforcement learning algorithm is used to estimate the action value of each commercial concrete shared weighbridge action performed on each grid in the grid area; An action-value map is determined based on the highest action value on each grid in the grid area.

[0010] In a possible implementation, the commercial concrete shared weighing scale actions include adding a new weighing scale, deleting a weighing scale, and moving a weighing scale.

[0011] In a possible implementation, a reinforcement learning algorithm is used based on the feature fusion graph to estimate the action value of each commercial concrete shared weighing scale action performed on each grid in the grid area, including: For each sub-time step, the reinforcement learning algorithm is used to calculate the action value of each commercial concrete shared weighbridge action. The action value includes the cost and reward of the commercial concrete shared weighbridge action. Among them, one commercial concrete shared weighbridge action can be executed in each sub-time step.

[0012] In a possible implementation, the site selection of a commercial concrete shared weighbridge is performed based on the commercial concrete shared weighbridge site selection action with the highest action value in the action value graph, including: Determine the target commercial concrete shared weighbridge location selection action with the highest action value in each grid in the action value graph; Execute the target commercial concrete shared weighbridge site selection action in the area corresponding to the grid.

[0013] The present invention also provides a commercial concrete shared weighbridge site selection device, comprising: An image making module is used to obtain the commercial concrete demand data and the weighbridge status data of the commercial concrete distribution area, and to draw the commercial concrete demand map and the weighbridge status map of the commercial concrete distribution area based on the commercial concrete demand data and the weighbridge status data; An action value calculation module is used to identify the commercial concrete demand graph and the weighbridge status graph using a preset dual-branch neural network model to obtain an action value graph, which is used to reflect the action value corresponding to the commercial concrete shared weighbridge site selection action in the commercial concrete distribution area; The site selection operation module is used to select the site of the commercial concrete shared weighbridge based on the commercial concrete shared weighbridge site selection action with the highest action value in the action value graph.

[0014] The present invention also provides an electronic device, comprising a memory and a processor, wherein: Memory, used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the method for selecting a site for a commercial concrete shared weighing scale in any of the above embodiments.

[0015] The present invention also provides a computer-readable storage medium for storing computer-readable programs or instructions. When the programs or instructions are executed by a processor, the steps in the commercial concrete shared weighing scale site selection method in any of the above-mentioned embodiments can be implemented.

[0016] The beneficial effects of the present invention are as follows: the method for site selection of a commercial concrete shared scale provided by the present invention converts the commercial concrete demand data and the scale status data of the commercial concrete distribution area into a commercial concrete demand graph and a scale status graph, and adopts a double-branch neural network model to identify the commercial concrete demand graph and the scale status graph, thereby obtaining an action value graph for reflecting the action value corresponding to the site selection action of the commercial concrete shared scale in the commercial concrete distribution area, and performs site selection of the commercial concrete shared scale according to the commercial concrete shared scale site selection action with the highest action value. The method can consider complex commercial concrete site selection factors, respond to changing commercial concrete demand in real time, provide efficient site selection strategies, reduce energy consumption and unreasonable consumption in commercial concrete transportation, and save costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A schematic diagram of a process flow of a method for selecting a site for a commercial concrete shared weighbridge provided in an embodiment of the present invention; Figure 2 A flowchart of an implementation method of S101 provided in an embodiment of the present invention; Figure 3 A schematic flow chart of an implementation method of S102 provided in an embodiment of the present invention; Figure 4 A schematic diagram of the structure of a dual-branch neural network model provided by an embodiment of the present invention; Figure 5 A schematic diagram of a flow chart of a method for determining an action value graph provided by an embodiment of the present invention; Figure 6 A schematic diagram of a flow chart of a weighbridge site selection operation method provided by an embodiment of the present invention; Figure 7 A schematic diagram of the structure of a device provided by an embodiment of the present invention; Figure 8 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of the present invention and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0020] Reference to an "embodiment" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0021] In the embodiment of the present invention, in order to express the problem concisely, the symbols required to define the problem are first agreed upon as follows: :Construction site set, for , where x and y are the horizontal and vertical coordinates of the construction site address, and The construction site's start time and stop time are respectively. The construction site's demand for commercial concrete can only occur at and between; : Ready-mixed concrete plant and station collection, including all ready-mixed concrete plants and stations that can supply any construction site. It is the production site of ready-mixed concrete and the starting point of ready-mixed concrete transportation; : a collection of time steps, ; : The set of all possible locations for weighbridges is the candidate set of weighbridge locations. It includes suitable locations inside or next to the construction site and all other possible locations. When defining this set, we will try to include all possible locations to provide sufficient optimization space for the algorithm. : The set of all weighbridges that remain open (active) at time step t. The algorithm will choose to open some weighbridges and close others at certain time steps. When a weighbridge is opened and before it is closed, it is active and belongs to this set; : Requirement set, for ,in, , , , representing the construction site p In the t At this time step, there is a demand for commercial concrete, which requires a commercial concrete truck to be transported from the plant station. s One transportation operation and one order from the construction site may generate multiple demands r , it is necessary to pre-process the input data; W: Waybill Set, , for a , , a waybill will be generated immediately ,in, , ,and , Yes the nearest weighbridge; : Metric function, in meters, for all , calculate the distances between all possible locations involved based on the road network; : The cost of adding a weighbridge; : The cost of moving a weighbridge; : It is a real number, which is a penalty factor for weighing credibility. The farther the weighing scale is from the construction site, the less credible the weighing will be. This factor is greater than 1, which is used to emphasize the importance of the distance between the weighing scale and the construction site. : The newly opened scale set at the tth time step; : The scale transfer set at the tth time step, for ,have , .

[0022] Commercial concrete shared weighbridge site selection problem: Given , for t=123...T, facing the waybill set at the tth time step , find the optimal action , , to minimize the following objective:

[0023] in, It is the sum of the distances from the factory station to the weighbridge that all waybills have traveled, aiming to optimize the total transportation distance to achieve the goal of reducing transportation costs and energy conservation and emission reduction. It is the weight credibility penalty factor multiplied by the sum of the distances from the weighbridge to the construction site for all waybills. It is used to constrain the weighbridge to be not too far from the construction site to ensure the weight credibility. and They are the sum of the costs of setting up a new weighbridge and relocating a weighbridge, ensuring that the existing weighbridge can be effectively utilized.

[0024] A specific embodiment of the present invention discloses a method for selecting a site for a commercial concrete shared weighbridge, comprising: S101, obtaining commercial concrete demand data and weighbridge status data of a commercial concrete distribution area, and drawing a commercial concrete demand graph and a weighbridge status graph of the commercial concrete distribution area based on the commercial concrete demand data and weighbridge status data; S102, using a preset dual-branch neural network model to identify the commercial concrete demand graph and the weighbridge status graph to obtain an action value graph, which is used to reflect the action value corresponding to the commercial concrete shared weighbridge site selection action in the commercial concrete distribution area; S103, selecting a site for a commercial concrete shared weighbridge based on the commercial concrete shared weighbridge site selection action with the highest action value in the action value graph.

[0025] In the embodiment of the present invention, the commercial concrete distribution area refers to an area that has commercial concrete plants, construction sites, and commercial concrete shared scales, and requires site planning for commercial concrete shared scales. In the commercial concrete distribution area, commercial concrete distribution demand will be generated, and the commercial concrete demand data includes the construction sites that need commercial concrete, the demand for commercial concrete, the commercial concrete plants that provide commercial concrete, and the commercial concrete shared scales that weigh the commercial concrete. The scale status data includes the status of all commercial concrete shared scales in the commercial concrete distribution area, such as operating status, maintenance status, pending transfer status, stop status, etc. The commercial concrete demand diagram and the scale status diagram are respectively used to display the commercial concrete demand and scale status in the commercial concrete distribution area, and the specific drawing method will be described in detail later in the present invention.

[0026] In an embodiment of the present invention, a preset dual-branch neural network model can identify the commercial concrete demand graph and the weighbridge status graph, and output an action value graph. The action value graph is used to reflect the action value corresponding to the site selection action of the commercial concrete shared weighbridge in the commercial concrete distribution area. Based on the action value graph, the site selection of the commercial concrete shared weighbridge can be carried out.

[0027] The method for site selection of a commercial concrete shared weighbridge provided by the present invention converts the commercial concrete demand data and weighbridge status data of a commercial concrete distribution area into a commercial concrete demand graph and a weighbridge status graph, and adopts a double-branch neural network model to identify the commercial concrete demand graph and the weighbridge status graph, thereby obtaining an action value graph for reflecting the action value corresponding to the site selection action of the commercial concrete shared weighbridge in the commercial concrete distribution area, and performs site selection of the commercial concrete shared weighbridge according to the commercial concrete shared weighbridge site selection action with the highest action value. The method can consider complex commercial concrete site selection factors, respond to changing commercial concrete demand in real time, provide efficient site selection strategies, reduce energy consumption and unreasonable consumption in commercial concrete transportation, and save costs.

[0028] As a possible implementation mode of the present invention, in this implementation mode, Figure 2 As shown, the commercial concrete demand data and the weighbridge status data of the commercial concrete distribution area are obtained, and the commercial concrete demand graph and the weighbridge status graph of the commercial concrete distribution area are drawn based on the commercial concrete demand data and the weighbridge status data, including: S201, gridding the commercial concrete distribution area, and mapping the commercial concrete plant station, commercial concrete weighbridge and construction site in the commercial concrete distribution area into the grid area; S202, determining the waybill data of each construction site within a preset time step, and determining the pixel value of the grid corresponding to the construction site based on the waybill data to obtain a commercial concrete demand map; S203, determining the active state of each commercial concrete weighbridge, and determining the pixel value of the grid corresponding to the commercial concrete weighbridge based on the active state, to obtain a weighbridge state diagram.

[0029] In an embodiment of the present invention, the ready-mixed concrete distribution area can be gridded and divided into N×N grid areas, and then the geographical locations of all ready-mixed concrete plants, ready-mixed concrete shared scales and construction sites are extracted and mapped to the grid. Generally speaking, it is necessary to ensure that the side length of each grid is between 150 meters and 250 meters, which can reduce the action space while fine-grained modeling of the distribution area; further, the waybill data includes the address of the departure plant, the passing scale and the destination construction site. This step requires associating the waybill with the discretized plant and construction site, and placing the demand within a time step at the corresponding position of the construction site in the grid to construct a demand graph. A demand graph The size of is the same as the grid area size N×N of the commercial concrete distribution area, where the pixel value of each grid is the number of transportation behaviors generated by the construction site in time step t. Furthermore, since the algorithm involves continuously optimizing the layout of shared weighbridges in multiple consecutive time steps (i.e., in multiple consecutive days), it is necessary to maintain the active status of the weighbridges. The weighbridge status diagram The size of is the same as the grid area size N×N of the commercial concrete distribution area, where the pixel value of each grid is 0 or 1, 1 represents that the weighbridge is in an active state, and 0 represents that there is no weighbridge at this time. Based on this, the commercial concrete demand map and weighbridge status map of the commercial concrete distribution area can be constructed.

[0030] The embodiment of the present invention constructs a commercial concrete demand graph and a weighbridge status graph of a commercial concrete distribution area according to the commercial concrete demand data and the weighbridge status data, thereby facilitating the recognition of a subsequent neural network model.

[0031] As a possible implementation mode of the present invention, in this implementation mode, Figure 3 As shown in the figure, the preset dual-branch neural network model is used to identify the commercial concrete demand graph and the weighbridge status graph to obtain the action value graph, including: S301, using the first branch network of the preset double-branch neural network model to extract features from the commercial concrete demand graph to obtain a commercial concrete demand feature graph; S302, using the second branch network of the preset dual-branch neural network model to extract features from the weighbridge state diagram to obtain a weighbridge state feature diagram; S303, feature fusion is performed on the commercial concrete demand feature graph and the weighbridge status feature graph to obtain a feature fusion graph, and an action value graph is determined based on the feature fusion graph.

[0032] In the embodiment of the present invention, the two branches of the dual-branch neural network model respectively use continuous downsampling layers to extract high-order features of the environmental state, and obtain feature maps of different levels. Each feature map of the same level must be subjected to feature fusion by a gated feature fusion layer to determine the importance of commercial concrete demand features and weighbridge state features, and obtain fusion feature maps of different levels. Specifically, Figure 4 As shown in the figure, the downsampling layer consists of two consecutive convolutional layers connected to a maximum pooling layer. Taking the output of the first-level downsampling layer as an example, the process of the gated fusion mechanism of commercial concrete demand and weighbridge status features is as follows: For a high-level feature map extracted from a commercial concrete demand map And a high-order feature map extracted from a scale state map , first perform a linear transformation through a convolution operation: ,

[0033] in, represents the convolution operation, and is the convolution kernel, and is the bias term.

[0034] Next, the linearly transformed feature map and The gating signal is calculated by the activation function, which uses Sigmoid function:

[0035] in, for Sigmoid function, It is the gating weight of the element in the interval [0, 1], which controls the fusion ratio of the two feature maps.

[0036] According to the gate signal G, the commercial concrete demand graph and the weighbridge status graph are weighted and summed to obtain the fused feature graph. :

[0037] in, Represents a point-by-point multiplication operation, which means that through the gate signal right and Weight accordingly.

[0038] First-order fusion characteristic diagram of commercial concrete demand and weighbridge status Will be concatenated to the output of the first downsampling layer and , and then serve as the input of the second-level dual-branch downsampling [ , ]and[ , ] and repeat the above steps until all operations for each branch are completed.

[0039] In this embodiment of the present invention, the lower part of the network estimates the value of placing and removing a scale on each square, as well as the value of stopping the action (stopping the decision-making process at this time step) from the high-order features. The final output space size of the network is [ ; ; ], which consists of the action value map of adding a new weighbridge, the action value map of deleting a weighbridge, and the action value of stopping. This part uses successive upsampling layers, each of which consists of a deconvolution layer and two successive convolution layers. In order to consider whether the decision of this time step should be stopped from a global perspective, this network inputs the highest-order feature map output by the branch modeling part into a multi-layer perceptron to estimate the value of stopping the decision of this time step, and splices each level of the feature map of the branch modeling part into the action value estimation feature map of the same size, which can effectively preserve fine-grained information and supplement the detail information lost by the high-order features obtained by the convolutional neural network.

[0040] As a possible implementation mode of the present invention, in this implementation mode, Figure 5 As shown, the action value graph is determined based on the feature fusion graph, including: S501, using a reinforcement learning algorithm based on the feature fusion graph to estimate the action value of each commercial concrete shared weighbridge action executed on each grid in the grid area; S502, determining an action value graph based on the highest action value on each grid in the grid area.

[0041] In the embodiment of the present invention, when estimating the value of placing and deleting a weighbridge on each square or the value of moving the weighbridge step by step based on the feature fusion graph, it is necessary to optimize the estimation process in combination with a reinforcement learning algorithm. The specific reinforcement learning algorithm is as follows: enter( , ,P,S, ,T, , , Environment Action Modeling Network , exploration rate , experience replay cache M, discount factor , learning rate , target network update frequency C); output the trained environment action modeling network; initialize the experience storage cache M; initialize the network Parameters; Initialize target network parameters ; Define the path set R = reachable path (starting point, end point); for each training step: Choose actions according to the greedy strategy :

[0042] if If it is an end action or the number of sub-time steps of the day is greater than the total sub-time steps, the processing of the day is ended; Otherwise, execute , get reward and the next state ; Will Put it into the experience replay cache and randomly select a small batch from the experience replay library , calculate the target Q value;

[0043] Use gradient descent to minimize the loss function and update the network parameters :

[0044] Every C steps, the network parameters are Copied to the target network, and finally returned to the environment action modeling network parameters .

[0045] Furthermore, the commercial concrete shared scale actions include adding a new scale, deleting a scale, and moving a scale.

[0046] Furthermore, based on the feature fusion graph, a reinforcement learning algorithm is used to estimate the action value of each commercial concrete shared weighbridge action performed on each grid in the grid area, including: For each sub-time step, the reinforcement learning algorithm is used to calculate the action value of each commercial concrete shared weighbridge action. The action value includes the cost and reward of the commercial concrete shared weighbridge action. Among them, one commercial concrete shared weighbridge action can be executed in each sub-time step.

[0047] In the embodiment of the present invention, the reinforcement learning algorithm is based on the DQN algorithm and has a flexible sub-time step design. Considering that the installation and transportation of the shared weighbridge takes time to complete, one time step is defined as one day. However, considering that there may be 0 or more new demands in an area within one day, it may be necessary to make 0 or more weighbridge action decisions within one time step. For this reason, an end action is designed, that is, a day is divided into 0 to multiple sub-time steps, and each step determines the action of a weighbridge until the model believes that the decision of the day should be ended, and then the decision of the next day is entered. To maintain robustness, this technology sets an experience upper limit L for each sub-time step of each day, and the optimization target is reflected in the reward in the reinforcement learning training stage. reward In the calculation of , the specific calculation formula is:

[0048] in, is the order that needs to be processed on day t, and are the action sets of adding or transferring a weighbridge in the current sub-time step of day t. In particular, since only one action is performed in each sub-time step, and They will not be greater than 0 at the same time, and at most equal to 1. Based on this, the reward value of executing an action of adding a weighbridge, transferring a weighbridge, or deleting a weighbridge in each current state can be calculated, so that the deep learning algorithm can optimize it to obtain the best weighbridge location action.

[0049] Further, such as Figure 6 As shown in the figure, the site selection of the commercial concrete shared weighbridge is performed based on the commercial concrete shared weighbridge site selection action with the highest action value in the action value graph, including: S601, determining the target commercial concrete shared weighbridge location selection action with the highest action value in each grid in the action value graph; S602, executing the target commercial concrete shared weighbridge site selection action in the area corresponding to the grid.

[0050] In an embodiment of the present invention, based on the action value graph obtained in the aforementioned embodiment, each grid corresponds to an action value of executing a certain commercial concrete shared scale site selection action, the commercial concrete shared scale site selection action with the highest action value is determined as the target commercial concrete shared scale site selection action of the grid, and in the area corresponding to the grid, one of the new construction, transfer or deletion of the scale is performed based on the commercial concrete shared scale site selection action.

[0051] The method for site selection of a commercial concrete shared weighbridge provided by the present invention converts the commercial concrete demand data and weighbridge status data of a commercial concrete distribution area into a commercial concrete demand graph and a weighbridge status graph, and adopts a double-branch neural network model to identify the commercial concrete demand graph and the weighbridge status graph, thereby obtaining an action value graph for reflecting the action value corresponding to the site selection action of the commercial concrete shared weighbridge in the commercial concrete distribution area, and performs site selection of the commercial concrete shared weighbridge according to the commercial concrete shared weighbridge site selection action with the highest action value. The method can consider complex commercial concrete site selection factors, respond to changing commercial concrete demand in real time, provide efficient site selection strategies, reduce energy consumption and unreasonable consumption in commercial concrete transportation, and save costs.

[0052] In order to better implement the commercial concrete shared weighbridge site selection method in the embodiment of the present invention, based on the commercial concrete shared weighbridge site selection method, correspondingly, Figure 7 As shown, the embodiment of the present invention further provides a site selection device for a commercial concrete shared weighbridge, and the site selection device 700 for a commercial concrete shared weighbridge includes: An image making module 701 is used to obtain the commercial concrete demand data and the weighbridge status data of the commercial concrete distribution area, and to draw the commercial concrete demand map and the weighbridge status map of the commercial concrete distribution area based on the commercial concrete demand data and the weighbridge status data; The action value calculation module 702 is used to identify the commercial concrete demand graph and the weighbridge status graph using a preset dual-branch neural network model to obtain an action value graph, which is used to reflect the action value corresponding to the commercial concrete shared weighbridge site selection action in the commercial concrete distribution area; The site selection operation module 703 is used to select a site for a commercial concrete shared weighbridge based on the commercial concrete shared weighbridge site selection action with the highest action value in the action value graph.

[0053] The commercial concrete shared weighbridge site selection device 700 provided in the above embodiment can implement the technical solution described in the above commercial concrete shared weighbridge site selection method embodiment. The specific implementation principles of the above modules or units can refer to the corresponding contents in the above commercial concrete shared weighbridge site selection method embodiment, which will not be repeated here.

[0054] like Figure 8 As shown, the present invention also provides an electronic device 800. The electronic device 800 includes a processor 801, a memory 802 and a display 803. Figure 8 Only some components of the electronic device 800 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0055] In some embodiments, the processor 801 can be a central processing unit (CPU), a microprocessor or other data processing chip, which is used to run the program code or process data stored in the memory 802, such as the commercial concrete shared weighing scale site selection method in the present invention.

[0056] In some embodiments, the processor 801 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processor 801 may be local or remote. In some embodiments, the processor 801 may be implemented in a cloud platform. In some embodiments, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination thereof.

[0057] In some embodiments, the memory 802 may be an internal storage unit of the electronic device 800, such as a hard disk or memory of the electronic device 800. In other embodiments, the memory 802 may also be an external storage device of the electronic device 800, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the electronic device 800.

[0058] Furthermore, the memory 802 may include both an internal storage unit of the electronic device 800 and an external storage device. The memory 802 is used to store application software installed in the electronic device 800 and various data.

[0059] In some embodiments, the display 803 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 803 is used to display information of the electronic device 800 and to display a visual user interface. The components 801-803 of the electronic device 800 communicate with each other via a system bus.

[0060] In some embodiments, when the processor 801 executes the commercial concrete shared weighbridge site selection program in the memory 802, the following steps may be implemented: Obtain the commercial concrete demand data and weighbridge status data of the commercial concrete distribution area, and draw the commercial concrete demand graph and weighbridge status graph of the commercial concrete distribution area based on the commercial concrete demand data and weighbridge status data; The preset dual-branch neural network model is used to identify the commercial concrete demand graph and the weighbridge status graph to obtain an action value graph, which is used to reflect the action value corresponding to the commercial concrete shared weighbridge site selection action in the commercial concrete distribution area; The site selection for the commercial concrete shared weighing scale is performed based on the commercial concrete shared weighing scale site selection action with the highest action value in the action value graph.

[0061] It should be understood that: when the processor 801 executes the commercial concrete shared weighbridge site selection program in the memory 802, in addition to the above functions, other functions can also be implemented. For details, please refer to the description of the corresponding method embodiment above.

[0062] Furthermore, the embodiment of the present invention does not specifically limit the type of the electronic device 800 mentioned, and the electronic device 800 may be a portable electronic device such as a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, etc. Exemplary embodiments of portable electronic devices include but are not limited to portable electronic devices equipped with IOS, Android, Microsoft or other operating systems. The above-mentioned portable electronic devices may also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, the electronic device 800 may not be a portable electronic device, but a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0063] Correspondingly, an embodiment of the present invention also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, it can implement the steps or functions of the commercial concrete shared weighing scale site selection method provided in the above-mentioned method embodiments.

[0064] Those skilled in the art can understand that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0065] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for selecting a site for a commercial concrete shared weighbridge, characterized in that: include: Acquire commercial concrete demand data and weighbridge status data of a commercial concrete distribution area, and draw a commercial concrete demand graph and a weighbridge status graph of the commercial concrete distribution area based on the commercial concrete demand data and the weighbridge status data; The commercial concrete demand graph and the weighbridge status graph are identified by using a preset dual-branch neural network model to obtain an action value graph, which is used to reflect the action value corresponding to the commercial concrete shared weighbridge site selection action in the commercial concrete distribution area; The site selection for the commercial concrete shared weighing scale is performed based on the commercial concrete shared weighing scale site selection action with the highest action value in the action value graph.

2. The method for selecting a site for a commercial concrete shared weighbridge according to claim 1, characterized in that: The step of acquiring the commercial concrete demand data and the weighbridge status data of the commercial concrete distribution area, and drawing the commercial concrete demand graph and the weighbridge status graph of the commercial concrete distribution area based on the commercial concrete demand data and the weighbridge status data, comprises: Gridding the commercial concrete distribution area, and mapping the commercial concrete plant, commercial concrete weighbridge and construction site in the commercial concrete distribution area into the grid area; Determine the waybill data of each construction site within a preset time step, and determine the pixel value of the grid corresponding to the construction site based on the waybill data to obtain a commercial concrete demand map; The active state of each commercial concrete scale is determined, and the pixel value of the grid corresponding to the commercial concrete scale is determined based on the active state to obtain a scale state diagram.

3. The method for selecting a site for a commercial concrete shared weighbridge according to claim 2, characterized in that: The preset dual-branch neural network model is used to identify the commercial concrete demand graph and the weighbridge state graph to obtain an action value graph, including: Using the first branch network of the preset dual-branch neural network model to extract features from the commercial concrete demand graph to obtain a commercial concrete demand feature graph; The second branch network of the preset dual-branch neural network model is used to extract features from the weighbridge state diagram to obtain a weighbridge state feature diagram; The commercial concrete demand characteristic graph and the weighbridge status characteristic graph are subjected to feature fusion to obtain a feature fusion graph, and an action value graph is determined based on the feature fusion graph.

4. The method for selecting a site for a commercial concrete shared weighbridge according to claim 3 is characterized in that: The determining of the action value graph based on the feature fusion graph includes: Based on the feature fusion graph, a reinforcement learning algorithm is used to estimate the action value of each commercial concrete shared weighing scale action executed on each grid in the grid area; An action value map is determined based on the highest action value on each grid in the grid area.

5. The method for selecting a site for a commercial concrete shared weighbridge according to claim 4, characterized in that: The commercial concrete shared weighing scale actions include adding a new weighing scale, deleting a weighing scale, and moving a weighing scale.

6. The method for selecting a site for a commercial concrete shared weighbridge according to claim 4, characterized in that: The method of estimating the action value of each commercial concrete shared weighing scale action performed on each grid in the grid area by using a reinforcement learning algorithm based on the feature fusion graph includes: For each sub-time step, a reinforcement learning algorithm is used to calculate the action value of each commercial concrete sharing weighing scale action, and the action value includes the cost and reward of the commercial concrete sharing weighing scale action, wherein one commercial concrete sharing weighing scale action can be executed in each sub-time step.

7. The method for selecting a site for a commercial concrete shared weighbridge according to claim 4, characterized in that: The site selection of the commercial concrete shared weighbridge based on the commercial concrete shared weighbridge site selection action with the highest action value in the action value graph comprises: Determine the target commercial concrete shared weighbridge location selection action with the highest action value in each grid in the action value graph; The target commercial concrete shared weighing scale site selection action is performed in the area corresponding to the grid.

8. A site selection device for a commercial concrete shared weighbridge, characterized in that: include: An image making module is used to obtain the commercial concrete demand data and the weighbridge status data of the commercial concrete distribution area, and draw the commercial concrete demand map and the weighbridge status map of the commercial concrete distribution area based on the commercial concrete demand data and the weighbridge status data; An action value calculation module, used to identify the commercial concrete demand graph and the scale state graph using a preset dual-branch neural network model to obtain an action value graph, wherein the action value graph is used to reflect the action value corresponding to the commercial concrete shared scale site selection action in the commercial concrete distribution area; A site selection operation module is used to select a site for a commercial concrete shared weighbridge based on the commercial concrete shared weighbridge site selection action with the highest action value in the action value graph.

9. An electronic device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the method for selecting a site for a commercial concrete shared weighing scale as described in any one of claims 1 to 7 above.

10. A computer-readable storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the method for selecting a site for a commercial concrete shared weighing scale as described in any one of claims 1 to 7.

Images