A palletizing path optimization control method for gravity energy storage system

Abstract

The present invention provides a palletizing path optimization control method for a gravity energy storage system, and relates to the field of palletizing path optimization control technology. The method comprises: obtaining the layout information of the stack in the gravity energy storage system, including the number of columns and rows of the stack, and the initial position and target position information of each gravity energy storage block; constructing a matrix model to represent the state of the stack according to the stack layout information, and determining the positioning information of the incoming train, including the position and posture of the train near the stack; constructing a finite element analysis model to simulate the real palletizing scene, and in the finite element analysis model, by simulating different palletizing paths, evaluating the efficiency and safety of each path to obtain the final palletizing path. The present invention optimizes the palletizing path through finite element analysis, improves the palletizing efficiency and safety, and realizes real-time monitoring, ensuring that the gravity energy storage blocks complete the palletizing accurately and efficiently.

Description

Technical Field

[0001] The present invention relates to the technical field of palletizing path optimization control, and in particular to a palletizing path optimization control method for a gravity energy storage system. Background Art

[0002] Traditional technologies rely on preset palletizing paths, some of which cannot be adjusted dynamically in complex and changing working environments. When encountering obstacles or emergencies, some lack the ability to respond in real time and re-plan the path, which affects production efficiency.

[0003] Due to the rigidity of path planning, when faced with goods of different types, sizes and weights, some cannot be stacked efficiently and accurately, resulting in errors in the stacking process, which in turn affects the stability and efficiency of the gravity energy storage system. In addition, in a complex stacking environment, the inability to timely identify and avoid potential safety risks, such as collisions, falls and other accidents, may also cause damage to operators and equipment, and even lead to more serious safety accidents. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a palletizing path optimization control method for a gravity energy storage system, thereby realizing efficient, safe and intelligent control of the palletizing process in the gravity energy storage system.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] A palletizing path optimization control method for a gravity energy storage system, the method comprising:

[0007] Obtaining the layout information of the stack in the gravity energy storage system, including the number of columns and rows of the stack and the initial position and target position information of each gravity energy storage block;

[0008] Based on the stack layout information, a matrix model is constructed to represent the state of the stack and determine the positioning information of the incoming train, including the position and posture of the train near the stack;

[0009] Construct a finite element analysis model to simulate the real palletizing scenario. In the finite element analysis model, by simulating different palletizing paths, evaluate the efficiency and safety of each path to obtain the final palletizing path;

[0010] Determine the number of gravity energy storage blocks to be grabbed at one time according to the number of train sections, control the fixture to grab the gravity energy storage blocks from the train according to the determined number, and place the gravity energy storage blocks at one time to the designated position in the stack along the final stacking path determined by the finite element analysis model;

[0011] During the palletizing operation, the situation on site is monitored in real time until all gravity energy storage blocks on the train are placed into the stack.

[0012] Furthermore, based on the stack layout information, a matrix model is constructed to represent the state of the stack and determine the positioning information of the incoming train, including the position and posture of the train near the stack, including:

[0013] Initialize a two-dimensional matrix model according to the stack layout information;

[0014] Traverse each gravity energy storage block in the stack, and set a mark at the corresponding position of the two-dimensional matrix according to its position information;

[0015] When the train enters the station and stops stably, the current position coordinates of the train are detected by the positioning system on the train, and the attitude sensor is used to measure and record the train's attitude information, including the train's tilt angle and direction;

[0016] The position coordinates and attitude data of the train are combined with the two-dimensional matrix model so that the fixture can pick up the gravity energy storage block from the train and place it in the correct position in the stack according to the predetermined path.

[0017] Furthermore, a finite element analysis model is constructed to simulate the actual palletizing scenario. In the finite element analysis model, different palletizing paths are simulated to evaluate the efficiency and safety of each path to obtain the final palletizing path, including:

[0018] Define a loss function that reflects the efficiency and safety of the palletizing path;

[0019] In the finite element analysis model, the path is represented by a series of parameters, including the coordinates, velocity and acceleration of key points, and a series of parameters as the starting point;

[0020] For each set of path parameters, the moving process of the fixture to place the gravity energy storage block into the stack along the path is simulated to obtain the results of finite element simulation;

[0021] Using the results of finite element simulation, the gradient of the loss function with respect to the path parameters is calculated;

[0022] According to the gradient of the path parameters, the path parameters are updated until the preset number of iterations is reached. When the iterative optimization is completed, the final palletizing path is output.

[0023] Furthermore, the loss function The calculation formula is:

[0024] ;

[0025] in, Indicates the total distance the fixture moves during the entire palletizing process, in meters; Indicates the time required to complete an operation, usually in seconds; It indicates the amount of work completed per unit time;

[0026] Indicates The tilt angle during the operation; Indicates the maximum safe tilt angle; Indicates Speed ​​during the operation; Indicates the maximum safe speed; Indicates the total time required to complete the entire palletizing operation; , and Represents the weight coefficient.

[0027] Further, according to the gradient of the path parameter, the path parameter is updated, including:

[0028] According to the gradient of the path parameter, Update path parameters;

[0029] in, Indicates the updated path parameters; Represents the path parameters in the current iteration; represents the initial learning rate; represents the cumulative sum of the squared gradients; represents a constant; Represents the gradient of the loss function with respect to the path parameters; Represents the coefficient, the value is (0, 1); Indicates the momentum value of the previous round; represents the regularization coefficient; where

[0030] ,in, ,

[0031] in, ;

[0032] in, and Respectively represent the number of the current iteration and the total number of iterations; Represents the total number of training samples; Indicates that all The training samples are summed; Indicates The target value of training samples; Indicates The predicted value of training samples; Indicates The feature vector of training samples; Represents the current path parameter vector.

[0033] Determine the number of gravity energy storage blocks to be grabbed at one time according to the number of train sections, control the fixture to grab the gravity energy storage blocks from the train according to the determined number, and place the gravity energy storage blocks at the designated position in the stack at one time along the final stacking path determined by the finite element analysis model, including:

[0034] According to the number of train sections, calculate the number of gravity energy storage blocks that need to be grabbed at one time;

[0035] The guiding fixture moves along a predetermined path to the position where the gravity energy storage block is located on the train;

[0036] When the clamp reaches the designated clamping position on the train, the clamping function of the clamp is activated so that the clamp clamps a predetermined number of gravity energy storage blocks;

[0037] The sensor integrated in the clamp confirms that the gravity energy storage block has been successfully clamped;

[0038] According to the final palletizing path, the fixture is guided to move to the designated position of the stack. During the movement, the fixture continuously detects the surrounding environment to identify potential obstacles.

[0039] When the fixture approaches the stack, it is positioned by sensors so that the gravity energy storage block is placed at the specified position in the stack to complete the palletizing process.

[0040] Furthermore, the number of rows and columns in the two-dimensional matrix corresponds to the number of rows and columns of the stack, respectively.

[0041] The above solution of the present invention includes at least the following beneficial effects:

[0042] By obtaining the layout information of the stack in the gravity energy storage system and constructing a matrix model to represent the state of the stack, the actual situation of the stack can be accurately grasped, providing accurate basic data for subsequent path planning. The finite element analysis model is used to simulate the real palletizing scene and evaluate the efficiency and safety of different paths, so that the optimal palletizing path can be selected, unnecessary movement and waiting time can be reduced, and the palletizing efficiency can be significantly improved. In the finite element analysis model, not only the efficiency of the path is considered, but also the safety of each path is fully evaluated. By simulating different palletizing paths, potential safety hazards such as collision, tilt, instability, etc. can be discovered and avoided in advance, thereby ensuring the safety of the palletizing process. Determining the number of energy storage blocks to be grabbed at one time according to the number of train sections can reasonably allocate resources and avoid waste or shortage of resources. At the same time, by accurately controlling the number of gravity energy storage blocks that the fixture grabs from the train and placing them at the designated position in the stack along the optimal path, the maximum utilization of resources can be achieved.

[0043] The use of automated control technology, through real-time monitoring of the on-site situation, and automatic adjustment and optimization according to preset algorithms and models, can reduce manual intervention and improve the automation level of the palletizing process. This not only reduces labor costs, but also improves the accuracy and stability of palletizing. It can be flexibly adjusted and optimized according to different stack layout information, the number of train sections, and the number and position of energy storage blocks. It has strong system adaptability and can perform well in both large-scale gravity energy storage systems and small experimental devices. The proposal and implementation of this method provides strong support for the further development of gravity energy storage technology. By optimizing the palletizing path and control method, the overall performance and economy of the gravity energy storage system can be improved, thereby promoting the application and promotion of gravity energy storage technology in a wider range of fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart of a method for optimizing control of a palletizing path for a gravity energy storage system provided by an embodiment of the present invention.

[0045] Figure 2 It is a flowchart of constructing a matrix model to represent the state of the stack according to the stack layout information and determining the positioning information of the incoming train, including the position and posture of the train near the stack in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0047] like Figure 1 to Figure 2 As shown, an embodiment of the present invention provides a palletizing path optimization control method for a gravity energy storage system, the method comprising the following steps:

[0048] Step 11, obtaining the layout information of the stack in the gravity energy storage system, including the number of columns and rows of the stack and the initial position and target position information of each gravity energy storage block;

[0049] Step 12, based on the stack layout information, construct a matrix model to represent the state of the stack and determine the positioning information of the incoming train, including the position and posture of the train near the stack;

[0050] Step 13, construct a finite element analysis model to simulate the actual palletizing scenario. In the finite element analysis model, by simulating different palletizing paths, the efficiency and safety of each path are evaluated to obtain the final palletizing path;

[0051] Step 14, determining the number of gravity energy storage blocks to be grabbed at one time according to the number of train sections, controlling the clamp to grab the gravity energy storage blocks from the train according to the determined number, and placing the gravity energy storage blocks at one time to the designated position in the stack along the final stacking path determined by the finite element analysis model;

[0052] Step 15, during the palletizing operation, monitor the on-site situation in real time until all the gravity energy storage blocks on the train are placed in the stack.

[0053] In an embodiment of the present invention, by obtaining detailed layout information of the stack in the gravity energy storage system, including the number of columns, the number of rows, and the initial position and target position information of each gravity energy storage block, the actual situation of the stack can be accurately grasped. The matrix model constructed according to the stack layout information concisely and clearly represents the state of the stack, which is convenient for the computer to process and analyze. At the same time, the positioning information of the incoming train, including the position and posture of the train near the stack, is determined, and a clear reference coordinate is provided for the subsequent palletizing operation, ensuring the accuracy of the palletizing. The constructed finite element analysis model can simulate the real palletizing scene, and by simulating different palletizing paths and comprehensively considering the efficiency and safety of the path, the optimal palletizing path is selected, which not only improves the palletizing efficiency, but also ensures the safety of the palletizing process and reduces potential safety hazards. The number of energy storage blocks to be grabbed at one time is determined according to the number of train sections, and the clamp is controlled to grab the gravity energy storage blocks from the train according to the determined number, realizing the intelligent grabbing operation. At the same time, along the final palletizing path determined by finite element analysis, the energy storage blocks are placed at the designated position in the stack at one time, improving the efficiency and accuracy of the palletizing operation. During the palletizing operation, by real-time monitoring of the on-site situation, possible problems can be discovered and dealt with in a timely manner, ensuring the smooth progress of the palletizing operation. At the same time, the monitoring data can also provide feedback for subsequent palletizing operations and further optimize the palletizing path and control strategy.

[0054] In a preferred embodiment of the present invention, the above step 11, obtaining the layout information of the stack in the gravity energy storage system, including the number of columns and rows of the stack and the initial position and target position information of each gravity energy storage block, may include:

[0055] Start the control software of the gravity energy storage system and ensure that the communication connection with the stack management system is normal.

[0056] Load the basic configuration information of the stack from the system database, including the total number of columns and rows of the stack, as well as the basic size and weight of each energy storage block. By interacting with the stack management system, send a request to obtain the number of columns and rows of the current stack. After receiving the request, the stack management system returns the corresponding number of columns and rows. The control software receives and parses this information and stores it in an appropriate data structure (such as a two-dimensional array or matrix) in the memory for subsequent use. Send a request to the stack management system to provide the initial position information of all gravity energy storage blocks in the current stack. Based on the request, the stack management system traverses each energy storage block in the stack and collects its position information (such as column number, row number, layer number, etc.), and returns the collected position information to the control software.

[0057] The control software receives and parses this information, creates a data object for each energy storage block, and stores the location information in these objects. According to the system's operating plan or the user's instructions, determine the target position to which each gravity energy storage block needs to be moved. The target position information may include the target column number, target row number, target layer number, etc. Store this target position information in the data object of the corresponding energy storage block, and verify the obtained number of columns, rows, initial position and target position information to ensure data integrity and consistency. If data abnormalities or missing data are found, communicate with the stack management system in a timely manner to request resending or supplementing data. If the stack changes (such as increasing or decreasing the number of columns or rows, or the energy storage block is moved, etc.), update the relevant data in the system in a timely manner to ensure that the system always has the latest status of the stack for accurate palletizing operations and control.

[0058] In a preferred embodiment of the present invention, the above step 12, based on the stack layout information, constructs a matrix model to represent the state of the stack and determines the positioning information of the incoming train, including the position and posture of the train near the stack, which may include:

[0059] Step 121, initializing a two-dimensional matrix model according to the stack layout information;

[0060] Step 122, traverse each gravity energy storage block in the stack, and set a mark at a corresponding position of the two-dimensional matrix according to its position information;

[0061] Step 123, when the train enters the station and stops stably, the current position coordinates of the train are detected by the positioning system on the train, and the attitude sensor is used to measure and record the attitude information of the train, including the tilt angle and direction of the train;

[0062] Step 124 , combining the position coordinates and posture data of the train with the two-dimensional matrix model, so that the fixture can grab the gravity energy storage block from the train and place it in the correct position in the stack according to the predetermined path.

[0063] In the embodiment of the present invention, according to the layout information of the stack (such as the number of rows), and the number of columns ), initialize a two-dimensional matrix , whose size is The initial value of all elements in the matrix is ​​set to 0, indicating that all positions in the stack are empty. Each gravity energy storage block in the stack is traversed, and for each energy storage block, the position information (such as row index) is used to calculate the energy storage block. and column index ),exist The corresponding position ( , ) (for example, set the element value to 1). This can intuitively indicate which positions in the stack are occupied and which positions are still free. When the train enters the station and stops stably, the positioning system on the train is started. The positioning system detects the current position coordinates of the train (such as , , ), these coordinates can be relative to the stack or a global reference frame. At the same time, the attitude sensor on the train is used to measure and record the train's attitude information, including the train's tilt angle (such as pitch angle, roll angle) and direction (such as yaw angle). The train's position coordinates and attitude data are compared with the previously initialized two-dimensional matrix Combined with the position and posture of the train, the optimal approach path and angle for the fixture to clamp the gravity energy storage block from the train are calculated. Based on the position and posture of the train and the idle state of the stack, a final path from the train to the stack is planned for the fixture. This path should avoid any obstacles and ensure that the fixture can accurately reach the energy storage block position on the train and the idle position in the stack.

[0064] Control the fixture to move along the planned path to the energy storage block position on the train. After the fixture arrives, adjust the angle and position of the fixture according to the train's posture information to ensure that the energy storage block can be firmly clamped. After successful clamping, the fixture moves along another planned path to an empty position in the stack. Before placing the energy storage block, the sensor is used again for precise calibration to ensure that the energy storage block can be accurately placed in the predetermined position. After successful placement, update To reflect the new status of the stack, continue to perform a series of operations from moving the fixture to the energy storage block position in the carriage, clamping the energy storage block, verifying successful clamping, planning and moving to the stack position, accurately calibrating, placing the energy storage block, and confirming successful placement, until the energy storage blocks in all carriages are completely grabbed and neatly stacked in the stack. During the whole process, the position, posture and status of the train are continuously monitored to ensure the accuracy and safety of the operation. When all energy storage blocks are successfully stacked in the stack, the operation is completed. You can choose to record the operation results (such as the number of energy storage blocks successfully grabbed, the stacking position, etc.) in the log or database.

[0065] By constructing a two-dimensional matrix model to represent the state of the stack, and combining the position and posture information of the train, the fixture can be accurately controlled to pick up the gravity energy storage block from the train and place it in the correct position in the stack. This improves the accuracy and efficiency of palletizing and reduces errors and repeated operations. Since the two-dimensional matrix model can be dynamically updated and modified, it can adapt to changes in different stack layouts and train positions. This enhances the flexibility and scalability of the system, enabling it to adapt to a variety of application scenarios and needs. Through automated and intelligent palletizing operations, manual intervention and labor intensity are reduced. At the same time, due to the improvement of palletizing accuracy and efficiency, operating costs and maintenance costs are also reduced. Real-time monitoring of the position and posture of the fixture, as well as the position and posture of the train, can promptly detect and deal with potential safety hazards. This improves the safety of the system and protects the safety of equipment and personnel.

[0066] In another preferred embodiment of the present invention, the number of rows and the number of columns in the two-dimensional matrix correspond to the number of rows and the number of columns of the stack, respectively, and may include:

[0067] Get the number of train carriages N and the matrix describing the layout of the energy storage blocks in each carriage In this matrix, Representative The specific layout of the energy storage blocks in the carriage, and the number of rows in the matrix and the number of columns Then they represent the row and column arrangement of the energy storage blocks inside the car. At the same time, we get the matrix describing the stack layout , the number of rows of the matrix and the number of columns Corresponding to the number of rows and columns of the stack respectively. In the initial state, all elements of this matrix are set to 0 to indicate that the stack is completely empty. Finally, a set of index variables are initialized, which will be used to track the number of the car currently being processed, the precise location of the energy storage block in the car, and the next available free position in the stack. Based on the position of the train, the layout of the energy storage block, and the stack position, the overall movement path of the fixture and the grasping and placement strategy are planned.

[0068] In a preferred embodiment of the present invention, the above step 13, constructing a finite element analysis model to simulate a real palletizing scenario, in which different palletizing paths are simulated in the finite element analysis model to evaluate the efficiency and safety of each path to obtain the final palletizing path, may include:

[0069] Step 131, defining a loss function reflecting the efficiency and safety of the palletizing path;

[0070] Step 132, in the finite element analysis model, the path is represented by a series of parameters, the series of parameters including the coordinates, velocity and acceleration of the key points, and the series of parameters is used as a starting point;

[0071] Step 133, step 133, for each set of path parameters, simulate the movement process of the fixture placing the gravity energy storage block into the stack along the path to obtain the result of finite element simulation;

[0072] Step 134, using the results of the finite element simulation, calculate the gradient of the loss function with respect to the path parameter;

[0073] Step 135, updating the path parameters according to the gradient of the path parameters until a preset number of iterations is reached, and when the iterative optimization is completed, outputting the final palletizing path.

[0074] In an embodiment of the present invention, a loss function that reflects the efficiency and safety of the palletizing path is defined. This function can comprehensively consider factors such as the length, time, energy consumption, and potential collision risk of the path. The specific form of the loss function can be adjusted and optimized according to actual needs, but it should be ensured that it can accurately reflect the pros and cons of the palletizing path. In the finite element analysis model, the palletizing path is represented as a series of parameters, including the coordinates, speed, and acceleration of key points. These parameters will serve as the starting point for path optimization and will be used for subsequent simulations and calculations. For each set of path parameters, the finite element analysis model is used to simulate the process of the fixture moving along the path and placing the gravity energy storage block into the stack. During the simulation process, factors such as the dynamic characteristics of the fixture, the weight and size of the gravity energy storage block, the layout of the stack, and possible obstacles need to be considered. The results of the simulation should include information such as the moving trajectory, speed, acceleration, and collision with the stack and other objects of the fixture. Using the results of the finite element simulation, the gradient of the loss function with respect to the path parameters is calculated. The gradient represents the sensitivity of the loss function to the change of the path parameters, which is the key information for path optimization. According to the calculated gradient information, the path parameters are updated using optimization algorithms (such as gradient descent method, Newton method, etc.). The updated path parameters should reduce the loss function value, that is, the efficiency and safety of the path are improved. Repeat the above simulation, calculation and update process until the preset number of iterations is reached or other stop conditions are met. In each iteration, the path parameters need to be adjusted and optimized to gradually approach the optimal solution. When the iterative optimization is completed, the final palletizing path is output. This path should be the final solution that takes into account multiple factors such as efficiency and safety, and can be used to guide actual palletizing operations.

[0075] By optimizing the palletizing path, the movement time and energy consumption of the fixture can be reduced, thereby improving palletizing efficiency.

[0076] The optimized path can also reduce waiting time and idle time, making the entire palletizing process smoother and more efficient. During the optimization process, factors such as potential collision risks and the dynamic characteristics of the fixture are taken into account, thereby enhancing the safety of the palletizing process. The optimized path can avoid collisions between the fixture and the stack or other objects, reducing the risk of accidents and damage. It can be adjusted and optimized according to factors such as different stack layouts, the size and weight of the gravity energy storage block, and the characteristics of the fixture. By optimizing the palletizing path, unnecessary movements and energy consumption can be reduced, thereby reducing operating costs and maintenance costs. At the same time, improving palletizing efficiency and safety can also reduce additional costs caused by accidents or damage.

[0077] In a preferred embodiment of the present invention, the loss function The calculation formula is:

[0078] ;

[0079] in, Indicates the total distance the fixture moves during the entire palletizing process, in meters; Indicates the time required to complete an operation, usually in seconds; It indicates the amount of work completed per unit time;

[0080] Indicates The tilt angle during the operation; Indicates the maximum safe tilt angle; Indicates Speed ​​during the operation; Indicates the maximum safe speed; Indicates the total time required to complete the entire palletizing operation; , and represents the weight coefficient; Indicates that all The sum is performed by the operation; is a "maximum value" function that compares 0 to size, and take the larger one; when hour, , indicating the The tilt angle of this operation exceeds the maximum safe tilt angle, which is an unsafe situation, so a corresponding penalty will be given in the loss function; when hour, , indicating the The tilt angle of this operation is within a safe range and will not affect the loss function. For comparison between 0 and When When The speed of this operation exceeds the maximum safe speed, which is a situation that needs to be avoided, so it will be reflected accordingly in the loss function; when When , it means that the speed is within a safe range and will not make additional contribution to the loss function.

[0081] In the embodiment of the present invention, the weight coefficient is set , and . Set the maximum safe tilt angle and maximum safe speed Initialize the total distance 0, total time =0, and variables for accumulating the tilt angle and speed excess. For each step in the palletizing process (e.g., each movement from the train picking up the gravity energy storage block to the stack):

[0082] Record the distance moved for each step and add it up to the total distance Record the time required for each step and add it up to the total time Get the inclination angle of the current step and speed . Calculate the excess tilt angle: If > , then calculate And added to the total of the excess tilt angle. > , then calculate And add it to the total of the excess speed. Use the formula Calculate the amount of work completed per unit time, where is the total time required to complete the entire palletizing operation (be careful to avoid division by zero). The loss function is calculated according to the formula .

[0083] pass In this term, the loss function encourages completing more work in a shorter time (i.e. moving a longer distance). This helps optimize the movement path and speed of the fixture, thereby reducing unnecessary pauses and waiting time and improving overall palletizing efficiency. The term is directly related to the time required to complete the entire palletizing operation. By minimizing this term, the loss function drives the system to find paths and strategies that can complete the task faster. In this term, the loss function penalizes operations that exceed the maximum safe tilt angle. This helps prevent instability or potential accidents caused by excessive tilting during the movement of the fixture, thereby enhancing the safety of the operation. Similarly, by In this term, the loss function penalizes operations that exceed the maximum safe speed. This helps prevent the gripper from moving too fast, which could cause collisions, loss of control, or other safety issues. By adjusting the weight coefficient , and , the loss function can be customized according to the specific palletizing tasks and requirements. For example, if you pay more attention to efficiency, you can increase and If you are more concerned about security, you can increase This flexibility allows the loss function to adapt to different application scenarios and requirements. The loss function provides gradient information about the path parameters, which is the key to path optimization. By calculating the gradient and using optimization algorithms (such as gradient descent, Newton's method, etc.), the path parameters can be iteratively updated to gradually approach the final solution. This iterative optimization process helps to continuously improve the efficiency and safety of the palletizing path.

[0084] In another preferred embodiment of the present invention, the above step 135 updates the path parameters according to the gradient of the path parameters, including: controlling the clamp to move along the planned path to the energy storage block position on the train. After the clamp arrives, the angle and position of the clamp are adjusted according to the posture information of the train to ensure that the energy storage block can be firmly clamped. After successful clamping, the clamp moves along another planned path to an empty position in the stack. Before placing the energy storage block, the sensor is used again for precise calibration to ensure that the energy storage block can be accurately placed in the predetermined position. After successful placement, update To reflect the new status of the stack, continue to perform a series of operations from moving the fixture to the energy storage block position in the carriage, clamping the energy storage block, verifying successful clamping, planning and moving to the stack position, accurately calibrating, placing the energy storage block, and confirming successful placement, until the energy storage blocks in all carriages are completely grabbed and neatly stacked in the stack. During the whole process, the position, posture and status of the train are continuously monitored to ensure the accuracy and safety of the operation. When all energy storage blocks are successfully stacked in the stack, the operation is completed. You can choose to record the operation results (such as the number of energy storage blocks successfully grabbed, the stacking position, etc.) in the log or database.

[0085] By constructing a two-dimensional matrix model to represent the state of the stack, and combining the position and posture information of the train, the fixture can be accurately controlled to pick up the gravity energy storage block from the train and place it in the correct position in the stack. This improves the accuracy and efficiency of palletizing and reduces errors and repeated operations. Since the two-dimensional matrix model can be dynamically updated and modified, it can adapt to changes in different stack layouts and train positions. This enhances the flexibility and scalability of the system, enabling it to adapt to a variety of application scenarios and needs. Through automated and intelligent palletizing operations, manual intervention and labor intensity are reduced. At the same time, due to the improvement of palletizing accuracy and efficiency, operating costs and maintenance costs are also reduced. Real-time monitoring of the position and posture of the fixture, as well as the position and posture of the train, can promptly detect and deal with potential safety hazards. This improves the safety of the system and protects the safety of equipment and personnel.

[0086] In another preferred embodiment of the present invention, the number of rows and the number of columns in the two-dimensional matrix correspond to the number of rows and the number of columns of the stack, respectively, and may include:

[0087] Get the number of train carriages N and the matrix describing the layout of the energy storage blocks in each carriage In this matrix, Representative The specific layout of the energy storage blocks in the carriage, and the number of rows in the matrix and number of columns Then they represent the row and column arrangement of the energy storage blocks inside the car. At the same time, we get the matrix describing the stack layout , the number of rows of the matrix and number of columns Corresponding to the number of rows and columns of the stack respectively. In the initial state, all elements of this matrix are set to 0 to indicate that the stack is completely empty. Finally, a set of index variables are initialized, which will be used to track the number of the car currently being processed, the precise location of the energy storage block in the car, and the next available free position in the stack. Based on the position of the train, the layout of the energy storage block, and the stack position, the overall movement path of the fixture and the grasping and placement strategy are planned.

[0088] In a preferred embodiment of the present invention, the above step 13, constructing a finite element analysis model to simulate a real palletizing scenario, in which different palletizing paths are simulated in the finite element analysis model to evaluate the efficiency and safety of each path to obtain the final palletizing path, may include:

[0089] Step 131, defining a loss function reflecting the efficiency and safety of the palletizing path;

[0090] Step 132, in the finite element analysis model, the path is represented by a series of parameters, including the coordinates, velocity and acceleration of the key points, and the series of parameters is used as the starting point;

[0091] Step 133, for each set of path parameters, simulate the process of the fixture moving along the path to place the gravity energy storage block into the stack to obtain the result of finite element simulation;

[0092] Step 134, using the results of the finite element simulation, calculate the gradient of the loss function with respect to the path parameter;

[0093] Step 135, updating the path parameters according to the gradient of the path parameters until a preset number of iterations is reached, and when the iterative optimization is completed, outputting the final palletizing path.

[0094] In an embodiment of the present invention, a loss function that reflects the efficiency and safety of the palletizing path is defined. This function can comprehensively consider factors such as the length, time, energy consumption, and potential collision risk of the path. The specific form of the loss function can be adjusted and optimized according to actual needs, but it should be ensured that it can accurately reflect the pros and cons of the palletizing path. In the finite element analysis model, the palletizing path is represented as a series of parameters, including the coordinates, speed, and acceleration of key points. These parameters will serve as the starting point for path optimization and will be used for subsequent simulations and calculations. For each set of path parameters, the finite element analysis model is used to simulate the process of the fixture moving along the path and placing the gravity energy storage block into the stack. During the simulation process, factors such as the dynamic characteristics of the fixture, the weight and size of the gravity energy storage block, the layout of the stack, and possible obstacles need to be considered. The results of the simulation should include information such as the moving trajectory, speed, acceleration, and collision with the stack and other objects of the fixture. Using the results of the finite element simulation, the gradient of the loss function with respect to the path parameters is calculated. The gradient represents the sensitivity of the loss function to the change of the path parameters, which is the key information for path optimization. According to the calculated gradient information, the path parameters are updated using optimization algorithms (such as gradient descent method, Newton method, etc.). The updated path parameters should reduce the loss function value, that is, the efficiency and safety of the path are improved. Repeat the above simulation, calculation and update process until the preset number of iterations is reached or other stop conditions are met. In each iteration, the path parameters need to be adjusted and optimized to gradually approach the optimal solution. When the iterative optimization is completed, the final palletizing path is output. This path should be the final solution that takes into account multiple factors such as efficiency and safety, and can be used to guide actual palletizing operations.

[0095] By optimizing the palletizing path, the movement time and energy consumption of the fixture can be reduced, thereby improving palletizing efficiency.

[0096] The optimized path can also reduce waiting time and idle time, making the entire palletizing process smoother and more efficient. During the optimization process, factors such as potential collision risks and the dynamic characteristics of the fixture are taken into account, thereby enhancing the safety of the palletizing process. The optimized path can avoid collisions between the fixture and the stack or other objects, reducing the risk of accidents and damage. It can be adjusted and optimized according to factors such as different stack layouts, the size and weight of the gravity energy storage block, and the characteristics of the fixture. By optimizing the palletizing path, unnecessary movements and energy consumption can be reduced, thereby reducing operating costs and maintenance costs. At the same time, improving palletizing efficiency and safety can also reduce additional costs caused by accidents or damage.

[0097] In a preferred embodiment of the present invention, the loss function The calculation formula is:

[0098] ;

[0099] in, Indicates the total distance the fixture moves during the entire palletizing process, in meters; Indicates the time required to complete an operation, usually in seconds; It indicates the amount of work completed per unit time;

[0100] Indicates The tilt angle during the operation; Indicates the maximum safe tilt angle; Indicates Speed ​​during the operation; Indicates the maximum safe speed; Indicates the total time required to complete the entire palletizing operation; , and Represents the weight coefficient.

[0101] In the embodiment of the present invention, the weight coefficient is set , and . Set the maximum safe tilt angle and maximum safe speed Initialize the total distance 0, total time =0, and variables for accumulating the tilt angle and speed excess. For each step in the palletizing process (e.g., each movement from the train picking up the gravity energy storage block to the stack):

[0102] Record the distance moved for each step and add it up to the total distance Record the time required for each step and add it up to the total time Get the inclination angle of the current step and speed . Calculate the excess tilt angle: If > , then calculate And added to the total of the excess tilt angle. > , then calculate And add it to the total of the excess speed. Use the formula Calculate the amount of work completed per unit time, where is the total time required to complete the entire palletizing operation (be careful to avoid division by zero). The loss function is calculated according to the formula .

[0103] pass In this term, the loss function encourages completing more work in a shorter time (i.e. moving a longer distance). This helps optimize the movement path and speed of the fixture, thereby reducing unnecessary pauses and waiting time and improving overall palletizing efficiency. The term is directly related to the time required to complete the entire palletizing operation. By minimizing this term, the loss function drives the system to find paths and strategies that can complete the task faster. In this term, the loss function penalizes operations that exceed the maximum safe tilt angle. This helps prevent instability or potential accidents caused by excessive tilting during the movement of the fixture, thereby enhancing the safety of the operation. Similarly, by In this term, the loss function penalizes operations that exceed the maximum safe speed. This helps prevent the gripper from moving too fast, which could cause collisions, loss of control, or other safety issues. By adjusting the weight coefficient , and , the loss function can be customized according to the specific palletizing tasks and requirements. For example, if you pay more attention to efficiency, you can increase and If you are more concerned about security, you can increase This flexibility allows the loss function to adapt to different application scenarios and requirements. The loss function provides gradient information about the path parameters, which is the key to path optimization. By calculating the gradient and using optimization algorithms (such as gradient descent, Newton's method, etc.), the path parameters can be iteratively updated to gradually approach the final solution. This iterative optimization process helps to continuously improve the efficiency and safety of the palletizing path.

[0104] In another preferred embodiment of the present invention, the above step 135, updating the path parameter according to the gradient of the path parameter, comprises: according to the gradient of the path parameter, by Update path parameters;

[0105] in, Indicates the updated path parameters; Represents the path parameters in the current iteration; represents the initial learning rate; represents the cumulative sum of the squared gradients; represents a constant; Represents the gradient of the loss function with respect to the path parameters; Represents the coefficient, the value is (0, 1); Indicates the momentum value of the previous round; represents the regularization coefficient; where

[0106] ,in, ,

[0107] in, ;

[0108] in, and Respectively represent the number of the current iteration and the total number of iterations; Represents the total number of training samples; Indicates that all The training samples are summed; Indicates The target value of training samples; Indicates The predicted value of training samples; Indicates The feature vector of training samples; Represents the current path parameter vector.

[0109] In the embodiment of the present invention, the initial path parameters are set . Set the initial learning rate . Set constant To avoid division by zero errors. Set the momentum coefficient , the value is in the range of (0,1). Set the regularization coefficient . Initialize the momentum value is a zero vector. Initialize the cumulative sum of the squared gradients is a zero vector. Set the number of iterations For each iteration (until the predetermined number of iterations is reached or other stopping conditions are met), the iteration counter is incremented. ← +1. According to the current path parameters And the loss function , calculate the gradient . Update the cumulative sum of squared gradients ← + Here, each component of the gradient vector is squared and accumulated. For each parameter component, the adaptive learning rate is calculated + .

[0110] Update momentum value Apply regularization The regularization term here is , but when updating we write it as To simplify the expression. Update the path parameters . Update the old value ; .

[0111] When the predetermined number of iterations is reached or other stopping conditions are met, the iteration ends. Output the updated path parameters. Output the final as the updated path parameter.

[0112] pass In this way, the learning rate can be calculated based on the cumulative sum of the squared gradients. Perform adaptive adjustments. In the early stages of training, the gradient is large and the learning rate is relatively large, which helps to converge quickly; as training progresses, the gradient gradually decreases and the learning rate also decreases accordingly, which helps to fine-tune the parameters and avoid oscillation and overfitting.

[0113] Introducing the momentum term , which can accelerate convergence and reduce oscillation. The momentum term is equivalent to a weighted average of the previous gradients, making the parameter update smoother and more stable. This term regularizes the path parameters, which helps prevent overfitting and improves the generalization ability of the model. The regularization term can penalize large parameter values, making the model more concise and robust. The gradient of the loss function with respect to the path parameters It is calculated by summing all training samples, which ensures the accuracy and stability of the gradient. Accurate gradient information helps the correctness and effectiveness of parameter updates. By introducing the constant , avoiding the situation where the denominator is zero and ensuring the stability of numerical calculations.

[0114] Determine the number of gravity energy storage blocks to be grabbed at one time according to the number of train sections, control the fixture to grab the gravity energy storage blocks from the train according to the determined number, and place the gravity energy storage blocks at one time to the designated position in the stack along the final stacking path determined by the finite element analysis model, which may include:

[0115] Step 145, calculating the number of gravity energy storage blocks that need to be grabbed at one time according to the number of train sections;

[0116] Step 146, guiding the fixture to move along a predetermined path to the position of the gravity energy storage block on the train;

[0117] Step 147, when the clamp reaches the designated clamping position on the train, activating the clamping function of the clamp so that the clamp clamps a predetermined number of gravity energy storage blocks;

[0118] Step 148, confirming that the gravity energy storage block has been successfully clamped through a sensor integrated on the clamp;

[0119] Step 149, according to the final palletizing path, the fixture is guided to move to the designated position of the stack. During the movement, the fixture continuously detects the surrounding environment to identify potential obstacles;

[0120] Step 150, when the fixture approaches the stack, positioning is performed through sensors so that the gravity energy storage block is placed at a designated position in the stack to complete the palletizing process.

[0121] In an embodiment of the present invention, the number of train sections is obtained. According to a preset rule or algorithm (such as a certain number of energy storage blocks corresponding to each train section), the number of energy storage blocks that need to be grabbed at one time is calculated. The calculated number of energy storage blocks is sent to the fixture control system. According to the position of the train and the storage position of the energy storage blocks, the moving path of the fixture is planned. The fixture is controlled to move along the predetermined path to the position of the gravity energy storage block on the train. The sensor confirms that the fixture has accurately reached the specified clamping position. A clamping instruction is sent to the fixture to activate its clamping function. The fixture clamps a predetermined number of gravity energy storage blocks according to the instruction. The sensor integrated in the fixture detects whether the energy storage block is successfully clamped. If it is detected that the energy storage block has been successfully clamped, the next step is continued; if it is unsuccessful, error processing is performed or the clamping is retried. According to the final stacking path, the moving path of the fixture from the train to the stack is planned. The fixture is controlled to move to the specified position of the stack along the planned path. During the movement, the fixture continuously detects the surrounding environment to identify potential obstacles and avoid collisions. When the fixture approaches the stack, it is precisely positioned by sensors. Based on the positioning information, a placement instruction is sent to the fixture. The fixture places the gravity energy storage block at the specified position in the stack according to the instruction. The sensor confirms that the energy storage block has been correctly placed, completing the palletizing process.

[0122] Suppose there is a train station with multiple carriages, each with a fixed number of gravity energy storage blocks at the back. These energy storage blocks need to be grabbed by an automated gripper and palletized into a nearby stack for subsequent use.

[0123] The operation process is as follows: Get the number of train carriages, assuming that the train has 5 carriages. The preset rule is that each carriage corresponds to 4 energy storage blocks. Therefore, the number of energy storage blocks that need to be grabbed at one time is 5 carriages × 4 / section = 20 energy storage blocks. Send the calculated number information of 20 energy storage blocks to the fixture control system. According to the position of the train and the storage location of the energy storage blocks (assuming it is at the end of the carriage), plan the movement path of the fixture. Control the fixture to move along the predetermined path to the location of the energy storage block at the end of the first carriage. Confirm through the sensor that the fixture has accurately reached the specified clamping position. Send a clamping instruction to the fixture to activate its clamping function. The fixture clamps 4 energy storage blocks (the number corresponding to the first carriage) according to the instruction. The sensor integrated in the fixture detects that the 4 energy storage blocks have been successfully clamped. According to the final stacking path, plan the movement path of the fixture from the current position to the stack. Control the fixture to move to the specified position of the stack along the planned path. During the movement, the fixture continuously detects the surrounding environment to ensure that there are no obstacles. When the fixture approaches the stack, it is precisely positioned by sensors. Based on the positioning information, a placement instruction is sent to the fixture. The fixture places the four energy storage blocks at the specified positions in the stack according to the instructions. The sensor confirms that the energy storage blocks have been correctly placed. Continue to perform a series of operations from moving the fixture to the energy storage block position in the carriage, clamping the energy storage block, verifying successful clamping, planning and moving to the stack position, precise calibration, placing the energy storage block, and confirming successful placement until the energy storage blocks in all carriages are completely grasped and neatly stacked in the stack.

[0124] Through automated fixtures and precise path planning, fast and accurate energy storage block grabbing and palletizing are achieved, improving work efficiency. The entire process requires almost no human intervention, reducing labor intensity, while also reducing errors and safety hazards caused by human factors. The fixture can continuously detect the surrounding environment during movement and clamping, identify and avoid potential obstacles, and ensure the safety of operation. Through sensors and precise control algorithms, it can ensure that the energy storage blocks are accurately grabbed and placed in the specified position, improving the accuracy of palletizing. It can be flexibly adjusted according to the number of train sections and the specifications of the energy storage blocks to adapt to different work scenarios and needs. The data in the entire process can be recorded and analyzed, providing strong support for subsequent optimization and improvement.

[0125] In a preferred embodiment of the present invention, the above step 15, during the palletizing operation, real-time monitoring of the on-site situation until all the gravity energy storage blocks on the train are placed in the stack, may include:

[0126] Activate a network of surveillance cameras and sensors at the palletizing operation site to ensure full coverage of the operation area.

[0127] Initialize the monitoring software, set monitoring parameters and alarm thresholds, so that alarms can be issued in time when abnormal situations occur. Start the fixture to perform palletizing operations according to the pre-planned path and strategy. Use the monitoring camera to observe the movement and grasping action of the fixture in real time to ensure that it operates according to the predetermined path and strategy. Use the sensor network to monitor the clamping force, movement speed and other key parameters to ensure the stability and accuracy of the operation. Monitor the position and posture information of the train in real time to ensure that the train remains stable during the palletizing process.

[0128] If the train position or posture changes, the stacking operation is immediately suspended and the cause is checked. The status of the stack is monitored in real time through monitoring cameras and sensors, including free positions, the number of energy storage blocks placed, etc.

[0129] Update the stack status matrix to reflect the latest palletizing situation. If any abnormal situation is found during monitoring (such as fixture failure, train movement, wrong stack position, etc.), an alarm will be immediately issued and the palletizing operation will be suspended. Analyze the cause of the abnormal situation and take appropriate treatment measures (such as repairing the fixture, repositioning the train, adjusting the stack position, etc.) until all gravity energy storage blocks on the train are successfully grabbed and placed in the stack.

[0130] After each pick and place operation, update the status information of the monitoring system and check if any abnormalities have occurred. When all energy storage blocks have been successfully palletized into the stack, stop the fixture operation and turn off the monitoring system. Check the palletizing results to ensure that all energy storage blocks are placed in the correct position and the stack status matrix is ​​consistent with the actual situation. Record the monitoring data, abnormal conditions and treatment measures during the palletizing operation for subsequent analysis and improvement. Generate a palletizing operation report, including information such as operation time, number of energy storage blocks successfully palletized, abnormal conditions and treatment results.

[0131] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A palletizing path optimization control method for a gravity energy storage system, characterized in that: The method comprises: Obtaining the layout information of the stack in the gravity energy storage system, including the number of columns and rows of the stack and the initial position and target position information of each gravity energy storage block; Based on the stack layout information, a matrix model is constructed to represent the state of the stack and determine the positioning information of the incoming train, including the position and posture of the train near the stack; Construct a finite element analysis model to simulate the real palletizing scenario. In the finite element analysis model, by simulating different palletizing paths, evaluate the efficiency and safety of each path to obtain the final palletizing path; Determine the number of gravity energy storage blocks to be grabbed at one time according to the number of train sections, control the fixture to grab the gravity energy storage blocks from the train according to the determined number, and place the gravity energy storage blocks at one time to the designated position in the stack along the final stacking path determined by the finite element analysis model; During the palletizing operation, the on-site situation is monitored in real time until all the gravity energy storage blocks on the train are placed in the stack; a finite element analysis model is built to simulate the real palletizing scenario. In the finite element analysis model, different palletizing paths are simulated to evaluate the efficiency and safety of each path to obtain the final palletizing path, including: Define a loss function that reflects the efficiency and safety of the palletizing path; In the finite element analysis model, the path is represented by a series of parameters, including the coordinates, velocity and acceleration of key points, and a series of parameters as the starting point; For each set of path parameters, the moving process of the fixture to place the gravity energy storage block into the stack along the path is simulated to obtain the results of finite element simulation; Using the results of finite element simulation, the gradient of the loss function with respect to the path parameters is calculated; According to the gradient of the path parameters, the path parameters are updated until the preset number of iterations is reached. When the iterative optimization is completed, the final palletizing path is output; Loss Function The calculation formula is: ； in, Indicates the total distance the fixture moves during the entire palletizing process, in meters; Indicates the time required to complete an operation, usually in seconds; It indicates the amount of work completed per unit time; Indicates The tilt angle during the operation; Indicates the maximum safe tilt angle; Indicates Speed ​​during the operation; Indicates the maximum safe speed; Indicates the total time required to complete the entire palletizing operation; , and Represents the weight coefficient.

2. The method for optimizing the palletizing path for a gravity energy storage system according to claim 1, characterized in that: According to the stack layout information, a matrix model is constructed to represent the state of the stack and determine the positioning information of the incoming train, including the position and posture of the train near the stack, including: Initialize a two-dimensional matrix model according to the stack layout information; Traverse each gravity energy storage block in the stack, and set a mark at the corresponding position of the two-dimensional matrix model according to its position information; When the train enters the station and stops stably, the current position coordinates of the train are detected by the positioning system on the train, and the attitude sensor is used to measure and record the train's attitude information, including the train's tilt angle and direction; The position coordinates and attitude data of the train are combined with the two-dimensional matrix model so that the fixture can pick up the gravity energy storage block from the train and place it in the correct position in the stack according to the predetermined path.

3. The method for optimizing the palletizing path for a gravity energy storage system according to claim 2, characterized in that: According to the gradient of the path parameter, update the path parameter, including: According to the gradient of the path parameter, Update path parameters; in, Indicates the updated path parameters; Represents the path parameters in the current iteration; represents the initial learning rate; represents the cumulative sum of the squared gradients; represents a constant; Represents the gradient of the loss function with respect to the path parameters; Represents the coefficient, the value is (0, 1); Indicates the momentum value of the previous round; represents the regularization coefficient; where ,in, , in, ; in, and Respectively represent the number of the current iteration and the total number of iterations; Represents the total number of training samples; Indicates that all The training samples are summed; Indicates The target value of training samples; Indicates The predicted value of training samples; Indicates The feature vector of training samples; Represents the current path parameter vector.

4. The method for optimizing the palletizing path for a gravity energy storage system according to claim 3, characterized in that: Determine the number of gravity energy storage blocks to be grabbed at one time according to the number of train sections, control the fixture to grab the gravity energy storage blocks from the train according to the determined number, and place the gravity energy storage blocks at the designated position in the stack at one time along the final stacking path determined by the finite element analysis model, including: According to the number of train sections, calculate the number of gravity energy storage blocks that need to be grabbed at one time; The guiding fixture moves along a predetermined path to the position where the gravity energy storage block is located on the train; When the clamp reaches the designated clamping position on the train, the clamping function of the clamp is activated so that the clamp clamps a predetermined number of gravity energy storage blocks; The sensor integrated in the clamp confirms that the gravity energy storage block has been successfully clamped; According to the final palletizing path, the fixture is guided to move to the designated position of the stack. During the movement, the fixture continuously detects the surrounding environment to identify potential obstacles. When the fixture approaches the stack, it is positioned by sensors so that the gravity energy storage block is placed at the specified position in the stack to complete the palletizing process.

5. The method for optimizing the palletizing path for a gravity energy storage system according to claim 4, characterized in that: The number of rows and columns in the two-dimensional matrix corresponds to the number of rows and columns in the stack, respectively.

Images