Tower bolt robot charging scheduling method and system

Abstract

The invention discloses a tower bolt robot charging scheduling method and system, and relates to the technical field of robot scheduling, and the method comprises the steps: collecting bolt operation, a robot state, a passage diagram, a charging seat state and power window data for a narrow channel of a tower and a scene with a small number of charging seats, and carrying out the alignment of the data into an alignment scheduling data set; the method comprises the following steps: predicting operation energy consumption by combining torque sampling and angle sampling, predicting waiting and passing energy consumption by superposing side capacity, generating a lower limit of safe electric quantity, segmenting charging time slots according to power windows, issuing reservation tokens, realizing window-based seating to reduce queuing, and cancelling the tokens and rearranging and scheduling when energy consumption deviation or time slot mismatch occurs. High-position retention and charging failure are reduced, and the maintenance window completion rate is improved.

Description

Technical Field

[0001] This invention relates to the field of robot scheduling technology, and in particular to a method and system for scheduling the charging of a tower bolt robot. Background Technology

[0002] Bolt tightening and inspection inside wind turbine towers are typically completed within the maintenance window. This involves limited working space, restricted access between ladders and platforms, and long retraction paths for robots after they ascend the tower, making passing difficult. On-site charging resources are often limited to a few charging stations at the tower base, and available charging power is constrained by factors such as auxiliary power supply, temperature rise, and contact resistance, leading to power window fluctuations and temporary unavailability of charging stations. Furthermore, differences in bolt condition can cause significant changes in the torque-angle curve, resulting in fluctuating or even sudden increases in operational energy consumption. If charging is only scheduled according to fixed thresholds or a first-come, first-served approach, problems such as insufficient power at higher elevations, wasted time queuing in access areas, mismatched charging time slots, and inability to complete tasks within the maintenance window on schedule can easily occur.

[0003] Currently, Chinese invention patent application number CN202510495328.3 discloses a method and apparatus for scheduling charging robots. This method acquires parking space data and vehicle data within the corresponding coverage area collected by data acquisition devices deployed on each charging robot used for charging vehicles. It then constructs a parking space distribution map based on the parking space data and a vehicle distribution map based on the vehicle data. When an idle charging robot exists, it schedules the idle charging robot to the target location corresponding to the idle parking space based on the parking space and vehicle distribution maps. This solution, when an idle charging robot exists, schedules it to the target location corresponding to the idle parking space in advance based on the parking space and vehicle distribution maps. After a vehicle is parked in an idle parking space and there is a charging demand, it can promptly schedule an idle charging robot, improving the scheduling efficiency of the charging robots.

[0004] The aforementioned technologies are insufficient to achieve reversible safety constraints for scheduled charging time slots under conditions of narrow tower passages, limited charging bases, fluctuating power windows, and uncertain energy consumption during bolt operations. Furthermore, they are insufficient to promptly revoke tokens and reschedule when energy consumption deviations or time slot mismatches occur, thus ensuring the completion of tasks within the maintenance window. Summary of the Invention

[0005] The technical problem solved by this invention is that the existing technology is difficult to achieve reversible safety constraints for scheduled charging time slots under conditions of narrow tower channels, limited charging bases, power window fluctuations, and uncertain energy consumption of bolting operations. It is also difficult to promptly revoke tokens and reschedule when energy consumption deviations or time slot mismatches occur to ensure the completion of tasks within the maintenance window.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] A method for scheduling the charging of a tower bolt robot includes the following steps:

[0008] Step S1: Collect tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data;

[0009] Step S2 involves cleaning, synchronizing, and mapping the tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data to form an aligned scheduling dataset.

[0010] Step S3: Calculate the task operation energy consumption prediction data and path travel energy consumption prediction data based on the aligned scheduling dataset, and output the task energy consumption prediction results.

[0011] Step S4: Calculate the minimum energy consumption data for rollback based on the task energy consumption prediction results and generate the safe power consumption lower limit data;

[0012] Step S5: Generate charging time slot data and issue charging reservation token data based on the safe power limit data and available charging power window data;

[0013] Step S6: Generate charging scheduling instructions based on charging reservation token data and task energy consumption prediction results and send them to the robot for execution; send back execution feedback data to update the alignment scheduling dataset.

[0014] Preferably, step S1 includes the following sub-steps:

[0015] Step S101: Collect tower bolt operation data, which includes bolt group identification, target torque data, torque sampling data, angle sampling data, operation start and end time data, and operation priority data.

[0016] Step S102: Collect robot status data, including SOC data, current position node data, walking speed data, task queue data, and battery capacity data.

[0017] Step S103: Collect tower passage map data, which includes node set data, edge set data, edge passage time data, and edge capacity data;

[0018] Step S104: Collect charging dock status data, which includes charging dock identification, maximum power data, temperature data, contact resistance data, and availability status data.

[0019] Step S105: Collect available charging power window data, which includes window start and end time data and corresponding available power upper limit data.

[0020] Preferably, step S2 includes the following sub-steps:

[0021] Step S201: Perform outlier removal and missing value interpolation on the tower bolt operation data, robot status data, and charging base status data to obtain cleaning data;

[0022] Step S202: Synchronize the cleaned data according to a unified timestamp to obtain synchronized data;

[0023] Step S203: Map the current location node data in the synchronization data to the passage map data, and unify the units of the edge passage time data to obtain the mapped data;

[0024] Step S204: Merge the mapping data and power window data by timestamp to form the aligned scheduling dataset.

[0025] Preferably, step S3 includes the following sub-steps:

[0026] Step S301: Calculate the mechanical work data of the bolt group operation based on the torque sampling data and the angle sampling data. The mechanical work data is the result obtained by curve integration of the torque sampling data and the angle sampling data.

[0027] Step S302: Calculate the task operation energy consumption prediction data based on the operation machinery power data, operation start and end time data and operation priority data, and output the task operation energy consumption prediction result;

[0028] Step S303: Calculate the predicted path travel time based on the tower passage map data, current location node data, and walking speed data;

[0029] Step S304: Calculate path waiting time prediction data based on the path travel time prediction data and the edge capacity data in the tower passage diagram data, and synthesize the path travel time prediction data with the path travel time prediction data to obtain path passage energy consumption prediction data.

[0030] Preferably, step S4 includes the following sub-steps:

[0031] Step S401: Based on the path passage energy consumption prediction data, solve the minimum energy consumption data for the robot to retreat from the current position node to the corresponding node of the nearest charging station on the tower passage map data;

[0032] Step S402: Convert the minimum energy consumption data and battery capacity data for pullback into pullback power demand data;

[0033] Step S403: Calculate congestion risk correction data based on edge capacity data and path waiting time prediction data;

[0034] Step S404: Overlay the withdrawn power demand data with the congestion risk correction data to generate a safe power lower limit data, and write the safe power lower limit data into the aligned scheduling dataset.

[0035] Preferably, step S5 includes the following sub-steps:

[0036] Step S501: parse the power window data and the maximum power data of the charging dock to generate the allocatable charging power data of each charging dock within the window;

[0037] Step S502: Based on the allocable charging power data, charge time slot data is generated by dividing it into segments according to a preset granularity. The charge time slot data includes charging dock identifier, time slot start and end time data, and time slot power data.

[0038] Step S503: Based on the safety power limit data, filter the robots that need to be recharged, match charging time slot data for each robot, and generate charging reservation token data;

[0039] Step S504: Send the charging reservation token data to the robot. The charging reservation token data includes arrival time window data, charging dock identifier, and timeout invalidation rule data.

[0040] Preferably, step S6 includes the following sub-steps:

[0041] Step S601: Using the task operation energy consumption prediction results, path passage energy consumption prediction data, charging time slot data and safe power limit data as inputs, construct rolling scheduling target data;

[0042] Step S602: Under the conditions of satisfying the edge capacity data constraints, the charging dock available state data constraints, and the safe power lower limit data constraints, the charging scheduling command is obtained by solving.

[0043] Step S603: The charging scheduling instruction and the charging reservation token data are merged to generate execution instruction data and issued. The execution instruction data includes task execution order data, path node sequence data and seated charging power data.

[0044] Preferably, step S6 further includes step S604:

[0045] Receive execution feedback data, which includes actual operation energy consumption data, actual path time data, charging dock temperature data, and contact resistance data;

[0046] Based on the execution feedback data, the energy consumption deviation data and time slot mismatch data are calculated. When the energy consumption deviation data or time slot mismatch data exceeds the preset threshold, the corresponding charging reservation token data is revoked and the charging time slot data is rematched to generate updated charging reservation token data and updated charging scheduling instructions.

[0047] Preferably, the update of the aligned scheduling dataset includes:

[0048] Replace the corresponding records in the task energy consumption prediction data with the actual operation energy consumption data, and correct the path waiting time prediction data with the actual path time data.

[0049] Write the charging dock temperature data and contact resistance data into the charging dock status data and update the charging dock available status data;

[0050] Based on the updated aligned scheduling dataset, the task operation energy consumption prediction data, safe power limit data, and charging time slot data are recalculated on a rolling basis to form the charging scheduling instruction for the next cycle. 1. A charging scheduling method for a tower bolt robot, comprising the following steps:

[0051] Step S1: Collect tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data;

[0052] Step S2 involves cleaning, synchronizing, and mapping the tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data to form an aligned scheduling dataset.

[0053] Step S3: Calculate the task operation energy consumption prediction data and path travel energy consumption prediction data based on the aligned scheduling dataset, and output the task energy consumption prediction results.

[0054] Step S4: Calculate the minimum energy consumption data for rollback based on the task energy consumption prediction results and generate the safe power consumption lower limit data;

[0055] Step S5: Generate charging time slot data and issue charging reservation token data based on the safe power limit data and available charging power window data;

[0056] Step S6: Generate charging scheduling instructions based on charging reservation token data and task energy consumption prediction results and send them to the robot for execution; send back execution feedback data to update the alignment scheduling dataset.

[0057] Preferably, step S1 includes the following sub-steps:

[0058] Step S101: Collect tower bolt operation data, which includes bolt group identification, target torque data, torque sampling data, angle sampling data, operation start and end time data, and operation priority data.

[0059] Step S102: Collect robot status data, including SOC data, current position node data, walking speed data, task queue data, and battery capacity data.

[0060] Step S103: Collect tower passage map data, which includes node set data, edge set data, edge passage time data, and edge capacity data;

[0061] Step S104: Collect charging dock status data, which includes charging dock identification, maximum power data, temperature data, contact resistance data, and availability status data.

[0062] Step S105: Collect available charging power window data, which includes window start and end time data and corresponding available power upper limit data.

[0063] Preferably, step S2 includes the following sub-steps:

[0064] Step S201: Perform outlier removal and missing value interpolation on the tower bolt operation data, robot status data, and charging base status data to obtain cleaning data;

[0065] Step S202: Synchronize the cleaned data according to a unified timestamp to obtain synchronized data;

[0066] Step S203: Map the current location node data in the synchronization data to the passage map data, and unify the units of the edge passage time data to obtain the mapped data;

[0067] Step S204: Merge the mapping data and power window data by timestamp to form the aligned scheduling dataset.

[0068] Preferably, step S3 includes the following sub-steps:

[0069] Step S301: Calculate the mechanical work data of the bolt group operation based on the torque sampling data and the angle sampling data. The mechanical work data is the result obtained by curve integration of the torque sampling data and the angle sampling data.

[0070] Step S302: Calculate the task operation energy consumption prediction data based on the operation machinery power data, operation start and end time data and operation priority data, and output the task operation energy consumption prediction result;

[0071] Step S303: Calculate the predicted path travel time based on the tower passage map data, current location node data, and walking speed data;

[0072] Step S304: Calculate path waiting time prediction data based on the path travel time prediction data and the edge capacity data in the tower passage diagram data, and synthesize the path travel time prediction data with the path travel time prediction data to obtain path passage energy consumption prediction data.

[0073] Preferably, step S4 includes the following sub-steps:

[0074] Step S401: Based on the path passage energy consumption prediction data, solve the minimum energy consumption data for the robot to retreat from the current position node to the corresponding node of the nearest charging station on the tower passage map data;

[0075] Step S402: Convert the minimum energy consumption data and battery capacity data for pullback into pullback power demand data;

[0076] Step S403: Calculate congestion risk correction data based on edge capacity data and path waiting time prediction data;

[0077] Step S404: Overlay the withdrawn power demand data with the congestion risk correction data to generate a safe power lower limit data, and write the safe power lower limit data into the aligned scheduling dataset.

[0078] Preferably, step S5 includes the following sub-steps:

[0079] Step S501: parse the power window data and the maximum power data of the charging dock to generate the allocatable charging power data of each charging dock within the window;

[0080] Step S502: Based on the allocable charging power data, charge time slot data is generated by dividing it into segments according to a preset granularity. The charge time slot data includes charging dock identifier, time slot start and end time data, and time slot power data.

[0081] Step S503: Based on the safety power limit data, filter the robots that need to be recharged, match charging time slot data for each robot, and generate charging reservation token data;

[0082] Step S504: Send the charging reservation token data to the robot. The charging reservation token data includes arrival time window data, charging dock identifier, and timeout invalidation rule data.

[0083] Preferably, step S6 includes the following sub-steps:

[0084] Step S601: Using the task operation energy consumption prediction results, path passage energy consumption prediction data, charging time slot data and safe power limit data as inputs, construct rolling scheduling target data;

[0085] Step S602: Under the conditions of satisfying the edge capacity data constraints, the charging dock available state data constraints, and the safe power lower limit data constraints, the charging scheduling command is obtained by solving.

[0086] Step S603: The charging scheduling instruction and the charging reservation token data are merged to generate execution instruction data and issued. The execution instruction data includes task execution order data, path node sequence data and seated charging power data.

[0087] Preferably, step S6 further includes step S604:

[0088] Receive execution feedback data, which includes actual operation energy consumption data, actual path time data, charging dock temperature data, and contact resistance data;

[0089] Based on the execution feedback data, the energy consumption deviation data and time slot mismatch data are calculated. When the energy consumption deviation data or time slot mismatch data exceeds the preset threshold, the corresponding charging reservation token data is revoked and the charging time slot data is rematched to generate updated charging reservation token data and updated charging scheduling instructions.

[0090] Preferably, the update of the aligned scheduling dataset includes:

[0091] Replace the corresponding records in the task energy consumption prediction data with the actual operation energy consumption data, and correct the path waiting time prediction data with the actual path time data.

[0092] Write the charging dock temperature data and contact resistance data into the charging dock status data and update the charging dock available status data;

[0093] Based on the updated aligned scheduling dataset, the task operation energy consumption prediction data, safe power limit data, and charging time slot data are recalculated in a rolling manner to form the charging scheduling instructions for the next cycle.

[0094] A charging scheduling system for a tower bolt robot includes a data acquisition module, a data alignment module, an energy consumption prediction module, an energy consumption calculation module, a charging reservation module, and a charging execution module.

[0095] The data acquisition module is used to collect tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data;

[0096] The data alignment module is used to clean, synchronize, and map the tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data to form an aligned scheduling dataset.

[0097] The energy consumption prediction module is used to calculate task operation energy consumption prediction data and path travel energy consumption prediction data based on the aligned scheduling dataset, and output the task energy consumption prediction result.

[0098] The energy consumption calculation module is used to calculate the minimum energy consumption data for rollback based on the task energy consumption prediction results and generate the safe power consumption lower limit data.

[0099] The charging reservation module is used to generate charging time slot data and issue charging reservation token data based on the safe power limit data and the available charging power window data.

[0100] The charging execution module is used to generate charging scheduling instructions based on charging reservation token data and task energy consumption prediction results, and send them to the robot for execution, and send back execution feedback data to update the alignment scheduling dataset.

[0101] The beneficial effects of this invention are as follows: This method is designed for scenarios with narrow passages in towers and a small number of charging stations. It collects data on bolt operations, robot status, traffic flow, charging station status, and power windows, and aligns them into an aligned scheduling dataset. It combines torque sampling and angle sampling to predict operational energy consumption, and overlays edge capacity to predict waiting and traffic energy consumption. It generates a safe lower limit for power consumption, divides charging time slots according to power windows, and issues reservation tokens to achieve window-based seating and reduce queuing. When energy consumption deviation or time slot mismatch occurs, the token is revoked and the scheduling is rearranged, reducing high-altitude congestion and charging failures, and improving the completion rate of maintenance windows. Attached Figure Description

[0102] Figure 1 A flowchart illustrating the steps of a tower bolt robot charging scheduling method according to an embodiment of the present invention;

[0103] Figure 2 This is a basic flowchart of a tower bolt robot charging scheduling system provided in one embodiment of the present invention. Detailed Implementation

[0104] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0105] Example 1, referring to Figure 1 A charging scheduling method for a tower bolt robot is provided, comprising the following steps:

[0106] Step S1: Collect tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data.

[0107] Step S2 involves cleaning, synchronizing, and mapping the tower bolt operation data, robot status data, tower access map data, charging base status data, and available charging power window data to form an aligned scheduling dataset.

[0108] Step S3: Calculate the task operation energy consumption prediction data and path travel energy consumption prediction data based on the aligned scheduling dataset, and output the task energy consumption prediction results.

[0109] Step S4: Calculate the minimum energy consumption data for rollback based on the task energy consumption prediction results and generate the safe power consumption lower limit data.

[0110] Step S5: Generate charging time slot data and issue charging reservation token data based on the safe power limit data and available charging power window data.

[0111] Step S6: Generate charging scheduling instructions based on charging reservation token data and task energy consumption prediction results and send them to the robot for execution; send back execution feedback data to update the alignment scheduling dataset.

[0112] This method targets scenarios with narrow tower passages and a limited number of charging stations. It collects data on bolt operations, robot status, traffic flow, charging station status, and power windows, aligning them into an aligned scheduling dataset. It combines torque and angle sampling to predict operational energy consumption, and overlays edge capacity to predict waiting and traffic energy consumption. It generates a safe lower limit for power consumption, divides charging slots according to power windows, and issues reservation tokens to enable window-based seating and reduce queuing. When energy consumption deviations or slot mismatches occur, the tokens are revoked and the scheduling is rearranged, reducing high-altitude congestion and charging failures, and improving the completion rate of maintenance windows.

[0113] Step S1 includes the following sub-steps:

[0114] Step S101: Collect tower bolt operation data. The tower bolt operation data includes bolt group identification, target torque data, torque sampling data, angle sampling data, operation start and end time data, and operation priority data.

[0115] When the robot is performing a bolt group operation, the control terminal generates a bolt group identifier for the bolt group and sends out the target torque data and operation priority data for the bolt group. The robot tightening actuator records the operation start and end time data at the start and end times of the operation.

[0116] Within the time interval corresponding to the start and end time data of the operation, the torque sensor and the angle encoder collect torque sampling data and angle sampling data at a fixed sampling period, and organize each sampling record into a triplet sequence of timestamp, torque value and angle value. The sampling period can be 50ms to 200ms, and each record carries a bolt group identifier for association.

[0117] Output tower bolt operation data (bolt group identification, target torque data, torque sampling data, angle sampling data, operation start and end time data, and operation priority data).

[0118] Step S102: Collect robot status data, including SOC data, current position node data, walking speed data, task queue data, and battery capacity data.

[0119] The robot battery management module periodically reports SOC data and battery capacity data, the robot positioning module outputs current position node data (corresponding to the node identifier in the node set of the passage map), the walking control module outputs walking speed data, and the control terminal maintains and sends out the robot's current queue of tasks to be executed (including the bolt group identifier sequence).

[0120] Output robot status data (SOC data, current position node data, walking speed data, pending task queue data, and battery capacity data).

[0121] Step S103: Collect tower passage map data, which includes node set data, edge set data, edge passage time data, and edge capacity data.

[0122] The platform inside the tower, the start and end points of the ladder section, and the location of the charging base are discretized into node set data; the passable segments between nodes are discretized into edge set data. For each edge, the edge passage time data (obtained by manual calibration) is recorded, and the edge capacity data (the maximum number of robots allowed to occupy and pass through at the same time) is recorded for each edge. The edge capacity data and the edge set data are stored in a one-to-one correspondence.

[0123] Output tower traffic map data (node ​​set data, edge set data, edge traffic time data, edge capacity data).

[0124] Step S104: Collect charging dock status data, which includes charging dock identification, maximum power data, temperature data, contact resistance data, and availability status data.

[0125] Each charging dock cycle uploads the charging dock identifier, maximum power data, temperature data, contact resistance data, and availability status data (available or unavailable). If the temperature data exceeds the preset temperature limit or the contact resistance data exceeds the preset contact resistance limit, the availability status data is set to unavailable.

[0126] Output charging dock status data (charging dock identifier, maximum power data, temperature data, contact resistance data, and availability data).

[0127] Step S105: Collect available charging power window data. The power window data includes window start and end time data and corresponding available power upper limit data.

[0128] The control unit obtains power window data within the future rolling window from the tower auxiliary power / maintenance power management side, including window start and end time data and corresponding available power upper limit data (such as providing segmented power upper limits in 5-minute increments).

[0129] Output power window data (window start and end time data and available power limit data).

[0130] Step S2 includes the following sub-steps:

[0131] Step S201: Outlier removal and missing value interpolation are performed on the tower bolt operation data, robot status data, and charging base status data to obtain cleaning data.

[0132] Outlier removal is performed on torque and angle sampling data: when a sudden change in torque sampling data exceeds a preset torque jump threshold and the duration is less than a preset minimum duration, it is judged as instantaneous noise and removed. Missing sampling points are linearly interpolated using the nearest neighboring values.

[0133] Missing values ​​were interpolated and obvious out-of-bounds values ​​were removed from the SOC data, temperature data, and contact resistance data in the same way.

[0134] Clean the data.

[0135] Step S202: Synchronize the cleaned data according to a unified timestamp to obtain synchronized data.

[0136] The cleaned data is aligned with a unified timestamp, and data from different sampling periods are mapped to a unified time raster using nearest neighbor or linear interpolation to obtain synchronized data.

[0137] Step S203: Map the current location node data in the synchronization data to the passage map data, and unify the units of the edge passage time data to obtain the mapped data.

[0138] The current location node data in the synchronized data is checked to see if it belongs to the node set data. If it does not belong, it is mapped to the nearest node according to the nearest node principle and is still output as the current location node data.

[0139] The edge passage time data will be standardized to the same unit, either seconds or milliseconds.

[0140] Output the mapping data.

[0141] Step S204: Merge the mapping data and power window data by timestamp to form an aligned scheduling dataset.

[0142] The mapping data and power window data are merged by timestamp to form an aligned scheduling dataset. The aligned scheduling dataset contains at least: bolt operation data, robot status data, traffic map data, charging dock status data, and power window data, and is queried in association under the same timestamp.

[0143] Step S3 includes the following sub-steps:

[0144] Step S301: Calculate the mechanical work data of the bolt group based on the torque sampling data and angle sampling data. The mechanical work data is the result obtained by curve integration of the torque sampling data and angle sampling data.

[0145] For each bolt group, the torque and angle sampling data are identified. Sampling points within the start and end time intervals of the operation are taken, sorted in ascending order of angle value, and the mechanical work data is calculated using trapezoidal integrals.

[0146] The angle difference between two adjacent points is The average torque is , The sampling point number is... For the first Torque data corresponding to each sampling point For the first The torque data corresponding to each sampling point are summed to obtain the working mechanical work data of the bolt group.

[0147] To avoid including alignment idling, a preset torque threshold is set: only sampling points where the torque value first reaches this threshold are included in the integration.

[0148] Step S302: Calculate the task operation energy consumption prediction data based on the operation machinery power data, operation start and end time data and operation priority data, and output the task operation energy consumption prediction result.

[0149] By combining the machine power data, start and end time data, and priority data, the predicted energy consumption data for this bolt group's task can be generated. Specifically, one of the following mapping methods can be used:

[0150] Linear mapping is used: the mechanical work data of the operation is converted into energy consumption by a preset conversion coefficient, and auxiliary consumption items related to the duration of the operation are added, and the output is still the predicted energy consumption data of the task operation.

[0151] The conversion coefficient can be calibrated using historical execution feedback data from the same robot and stored in the control terminal parameter table.

[0152] Output task energy consumption prediction data and task energy consumption prediction results.

[0153] Step S303: Calculate the predicted path travel time based on the tower passage map data, current location node data, and walking speed data.

[0154] Starting from the current position node data in the robot state data, and taking the corresponding node of the next bolt group and the corresponding node of the charging seat in the task queue data as candidate endpoints, the shortest travel time is searched based on the edge set data and the edge travel time data, and the path travel time prediction data of the corresponding path is output.

[0155] Step S304: Calculate path waiting time prediction data based on the path travel time prediction data and the edge capacity data in the tower passage diagram data, and synthesize the path travel time prediction data with the path travel time prediction data to obtain path passage energy consumption prediction data.

[0156] For each edge of the robot on the candidate path:

[0157] Based on the path travel time prediction data obtained in step S303, calculate the robot's estimated entry time and estimated exit time to that side:

[0158] The estimated entry time is the sum of the edge travel times accumulated from the starting point of the path to all edges preceding this edge;

[0159] The estimated departure time is the sum of the estimated entry time and the edge travel time data for that edge.

[0160] Within the same projected time window, count the number of robots expected to simultaneously occupy that side.

[0161] When the number of data occupied is less than or equal to the edge capacity, it is considered that no waiting is required, and the waiting time for that edge is 0.

[0162] When the number of data occupied exceeds the edge capacity, it is assumed that they will queue up and wait.

[0163] For the edge where excess occurs, let:

[0164] The excess quantity is the difference between the number of simultaneous occupancy and the edge capacity data. The waiting time of this edge is estimated by multiplying the excess quantity and the edge passage time data (that is, for each additional excess robot, there will be at least one more passage time). The waiting time of this edge is the product of the excess quantity and the edge passage time data. The waiting times of all edges on the candidate path are added together to obtain the path waiting time prediction data of the candidate path.

[0165] Calculate the predicted path travel time, which is the sum of the predicted path travel time and the predicted path waiting time. Two types of energy consumption conversion factors per unit time are preset at the control terminal:

[0166] Energy consumption conversion factor per unit time of movement (corresponding to movement consumption);

[0167] Energy consumption conversion factor per unit time while waiting (corresponding to standby power consumption).

[0168] but:

[0169] Mobile energy consumption is the product of the mobile unit time energy consumption conversion factor and the predicted route travel time data;

[0170] Waiting energy consumption is the product of the energy consumption conversion factor per unit waiting time and the path waiting time prediction data.

[0171] The two are added together to obtain the predicted energy consumption data for the route.

[0172] Step S4 includes the following sub-steps:

[0173] Step S401: Based on the path travel energy consumption prediction data, solve the minimum energy consumption data for the robot to retreat from the current position node to the corresponding node of the nearest charging station on the tower travel map data.

[0174] For each robot, starting from the current location node data, and using all available status data as the corresponding nodes of available charging docks as candidate endpoints, the minimum value is obtained on the traffic map data based on the path passage energy consumption prediction data, thus obtaining the minimum retreat energy consumption data.

[0175] Step S402: Convert the minimum energy consumption data and battery capacity data for pullback into pullback power demand data.

[0176] Divide the minimum drawdown energy consumption data by the battery capacity data to obtain the drawdown power demand data.

[0177] Step S403: Calculate congestion risk correction data based on edge capacity data and path waiting time prediction data.

[0178] For each edge on the robot's retreat path, extract the edge capacity data and the path waiting time prediction data to construct a congestion coefficient. :

[0179] ;

[0180] in, This is the edge capacity data for that edge. This is the preset correction factor.

[0181] Then, the congestion factor is applied to the path waiting time prediction data to obtain an additional waiting time margin. :

[0182] ;

[0183] in, This is the path waiting time prediction data.

[0184] The control terminal presets a power consumption coefficient per unit waiting time. Then the congestion risk correction data for:

[0185] ;

[0186] The logic for determining the power consumption coefficient per unit of waiting time is as follows:

[0187] Find a period of time spent waiting in place, and use the actual path time of that period as the waiting duration. Then take the difference in SOC data before and after the wait. .

[0188] ;

[0189] Step S404: Overlay the withdrawn power demand data with the congestion risk correction data to generate a safe power lower limit data, and write the safe power lower limit data into the aligned scheduling dataset.

[0190] The power demand data is overlaid with the congestion risk correction data to obtain the safe power lower limit data, which is then written into the aligned scheduling dataset for subsequent screening of robots that need to be recharged.

[0191] Step S5 includes the following sub-steps:

[0192] Step S501: parse the power window data and the maximum power data of the charging dock to generate the allocatable charging power data of each charging dock within the window.

[0193] For each power window data, the start and end time periods are:

[0194] The available power limit data is allocated according to the number of charging docks and the maximum power data to obtain the allocable charging power data for each charging dock within the window;

[0195] When the availability status data of a certain charging dock is unavailable, its allocable charging power data is set to 0, and the remaining power is allocated to other available charging docks according to the same rules.

[0196] Step S502: Based on the allocable charging power data, charge time slot data is generated by dividing it into segments according to a preset granularity. The charge time slot data includes the charging dock identifier, time slot start and end time data, and time slot power data.

[0197] The window start and end time period is divided according to a preset granularity (e.g., 1 minute) to form charging time slot data. Each time slot records the charging dock identifier, time slot start and end time data, and time slot power data.

[0198] Step S503: Based on the safety power limit data, select the robots that need to be recharged, match charging time slot data for each robot, and generate charging reservation token data.

[0199] For each robot, compare the SOC (State of Charge) data with the safe minimum charge data: when the SOC data is less than or equal to the safe minimum charge data, it is determined that it needs to be recharged.

[0200] For robots that need to be recharged, the earliest arrival time is estimated by combining the path travel time prediction data and the path waiting time prediction data. The time slot that meets the arrival time falling into the time slot start and end time data is selected from the charging time slot data. Priority is given to the charging slot corresponding to the charging seat where the available status data is available and the temperature data and contact resistance data are within the normal range.

[0201] Generate charging reservation token data, which includes at least: arrival time window data, charging station identifier, and timeout invalidation rule data. The timeout invalidation rule data can be set to invalidate the token if the user is not seated by the end of the arrival time window.

[0202] Step S504: Send the charging reservation token data to the robot. The charging reservation token data includes arrival time window data, charging dock identifier, and overdue invalidation rule data.

[0203] The charging reservation token data is sent to the corresponding robot, and the token is written into the alignment scheduling dataset for use in S6 scheduling solution and execution constraints.

[0204] Step S6 includes the following sub-steps:

[0205] Step S601: Using the task operation energy consumption prediction results, path passage energy consumption prediction data, charging time slot data, and safe power limit data as inputs, construct rolling scheduling target data.

[0206] Using the task operation energy consumption prediction results, route passage energy consumption prediction data, charging time slot data, safe power limit data, and charging reservation token data as inputs, rolling scheduling target data is constructed, and the rolling window length can be consistent with the prediction length of the power window data.

[0207] Step S602: Under the conditions of satisfying the edge capacity data constraints, the charging dock available state data constraints, and the safe power lower limit data constraints, the charging scheduling command is obtained.

[0208] Generate charging scheduling instructions under the following constraints:

[0209] Edge capacity data constraint: At any given time, the number of robots occupying the same edge does not exceed the edge capacity data;

[0210] Charging dock availability status data constraint: Only charging docks with available status data are allowed to be assigned the status of being available;

[0211] Safety power limit data constraint: The predicted power consumption of the robot during task execution and passage must not be lower than the safety power limit data;

[0212] Charging time slot data constraints: The start and end of charging must fall within the matched start and end time data of the time slot, and the power must not exceed the time slot power data.

[0213] The solution results should include at least whether each robot performs its task first or returns to charging first within the scrolling window, the corresponding charging dock identifier, and the corresponding time slot.

[0214] Step S603: The charging scheduling instruction and the charging reservation token data are merged to generate execution instruction data and issued. The execution instruction data includes task execution order data, path node sequence data and seated charging power data.

[0215] The charging scheduling instruction and the charging reservation token data are merged to generate execution instruction data, which is then issued. The execution instruction data includes:

[0216] Task execution order data (selected and rearranged from the queue of tasks to be executed);

[0217] Path node sequence data (obtained from the travel map data);

[0218] Power data for in-seat charging (not exceeding the power data in the time slot).

[0219] The robot moves according to the path node sequence data, takes its seat within the time window, and charges according to the charging power data. After charging is completed, it continues to execute tasks according to the task execution order data.

[0220] The robot moves according to the path node sequence and performs docking and token authentication upon reaching the charging dock. After successful authentication, it performs constant power charging according to the docking charging power data. After completion, it exits the charging dock and continues to perform tasks. When the energy consumption deviation or time slot mismatch exceeds the threshold, the control terminal cancels the charging reservation token and rematches the charging time slot, and at the same time sends updated execution command data, which the robot then executes accordingly.

[0221] Step S6 also includes step S604:

[0222] Receive execution feedback data, which includes actual operation energy consumption data, actual path time data, charging dock temperature data, and contact resistance data.

[0223] Based on the execution feedback data, the energy consumption deviation data and time slot mismatch data are calculated. When the energy consumption deviation data or time slot mismatch data exceeds the preset threshold, the corresponding charging reservation token data is revoked and the charging time slot data is rematched to generate updated charging reservation token data and updated charging scheduling instructions.

[0224] The control terminal receives execution feedback data, which includes:

[0225] Actual operational energy consumption data (summarized by bolt group identification);

[0226] Actual path time data (summary of walking segments corresponding to path node sequence data);

[0227] Charging stand temperature data and contact resistance data (summarized according to charging stand label).

[0228] Energy consumption deviation data and time slot mismatch data are calculated based on execution feedback data:

[0229] Energy consumption deviation data: The difference or ratio between actual operational energy consumption data and predicted task operational energy consumption data;

[0230] Time slot mismatch data: Compare the deviation between the actual arrival and seating time and the arrival time window data, as well as the deviation between the actual charging duration and the time slot start and end time data.

[0231] Set preset thresholds: The energy consumption deviation threshold can be set according to a fixed proportion of the task operation energy consumption prediction data; the time slot mismatch threshold can be set according to a fixed proportion of the arrival time window data length (the threshold value is not limited, but it must be configurable in the control terminal parameter table).

[0232] When energy consumption deviation data or time slot mismatch data exceeds a preset threshold:

[0233] Cancel the corresponding charging reservation token data (mark the token as invalid and release the charging time slot data it occupies);

[0234] Re-match the charging time slot data to generate updated charging reservation token data;

[0235] The updated charging reservation token data is used to re-solve and generate updated charging scheduling instructions;

[0236] The updated charging reservation token data and the updated charging scheduling instructions are sent to the corresponding robots for execution.

[0237] The updates to the aligned scheduling dataset include:

[0238] Replace the corresponding records in the task energy consumption prediction data with the actual operation energy consumption data, and correct the path waiting time prediction data with the actual path time data.

[0239] Write the charging dock temperature data and contact resistance data into the charging dock status data and update the charging dock availability status data.

[0240] Based on the updated aligned scheduling dataset, the task operation energy consumption prediction data, safe power limit data, and charging time slot data are recalculated in a rolling manner to form the charging scheduling instructions for the next cycle.

[0241] Example 2, refer to Figure 2 A charging scheduling system for a tower bolt robot is provided, including a data acquisition module, a data alignment module, an energy consumption prediction module, an energy consumption calculation module, a charging reservation module, and a charging execution module.

[0242] The data acquisition module is used to collect tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data.

[0243] The data alignment module is used to clean, synchronize, and map the tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data to form an aligned scheduling dataset.

[0244] The energy consumption prediction module is used to calculate task operation energy consumption prediction data and path travel energy consumption prediction data based on the aligned scheduling dataset, and output the task energy consumption prediction results.

[0245] The energy consumption calculation module is used to calculate the minimum energy consumption data for rollback based on the task energy consumption prediction results and generate the lower limit data for safe power consumption.

[0246] The charging reservation module is used to generate charging time slot data and issue charging reservation token data based on the safe power limit data and the available charging power window data.

[0247] The charging execution module is used to generate charging scheduling instructions based on charging reservation token data and task energy consumption prediction results, and send them to the robot for execution, and send back execution feedback data to update the alignment scheduling dataset.

[0248] This invention manages robot status, traffic map, and power window in a unified manner by aligning the scheduling dataset, and solves the minimum energy consumption data for retreat based on the path traffic energy consumption prediction data, and further generates safe power limit data, so that robot scheduling can meet the retreat requirement at any time, significantly reducing the risk of being stranded at high altitudes due to power outages and the need for manual rescue.

[0249] By introducing edge capacity data to calculate path waiting time prediction data, congestion waiting is explicitly included in energy consumption prediction and scheduling constraints, avoiding queuing and ineffective energy consumption caused by multiple robots competing for space on stair sections or narrow edges, and improving traffic efficiency.

[0250] Based on power window data and charging station status data, charging time slot data is generated. Then, charging reservation token data is used to enable arrival and seating by window, reducing long queues and wasted time slots caused by first-come, first-served, and enabling the limited number of charging stations to serve multiple robots more efficiently within the maintenance window.

[0251] By incorporating charging dock temperature data, contact resistance data, and availability status data into the time slot generation and matching rules, unavailable resources can be automatically avoided when the charging dock overheats or has abnormal contact. The time slots are also updated in a rolling manner as the power window changes, reducing the charging failure rate.

[0252] The system uses torque and angle sampling data to calculate the mechanical work data and generate task energy consumption prediction data, making the energy consumption prediction relevant to the actual state of the bolts. At the same time, it receives execution feedback data, calculates energy consumption deviation data and time slot mismatch data, triggers token revocation and rematch, and generates updated charging reservation token data and updated charging scheduling instructions to ensure that the scheduling scheme can still converge and execute under disturbances.

[0253] By writing actual operation energy consumption data and actual path time data back to the aligned scheduling dataset, the task operation energy consumption prediction and path waiting prediction can be continuously corrected, making subsequent scheduling more in line with the specific passage characteristics of the tower and the individual differences of the robot, thereby improving the overall operation throughput and window achievement rate.

[0254] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0255] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the protection scope of the present invention.

Claims

1. A charging scheduling method for a tower bolt robot, characterized in that, Includes the following steps: Step S1: Collect tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data; Step S2 involves cleaning, synchronizing, and mapping the tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data to form an aligned scheduling dataset. Step S3: Calculate the task operation energy consumption prediction data and path travel energy consumption prediction data based on the aligned scheduling dataset, and output the task energy consumption prediction results. Step S4: Calculate the minimum energy consumption data for rollback based on the task energy consumption prediction results and generate the safe power consumption lower limit data; Step S5: Generate charging time slot data and issue charging reservation token data based on the safe power limit data and available charging power window data; Step S6: Generate charging scheduling instructions based on charging reservation token data and task energy consumption prediction results and send them to the robot for execution; send back execution feedback data to update the alignment scheduling dataset.

2. The tower bolt robot charging scheduling method as described in claim 1, characterized in that, Step S1 includes the following sub-steps: Step S101: Collect tower bolt operation data, which includes bolt group identification, target torque data, torque sampling data, angle sampling data, operation start and end time data, and operation priority data. Step S102: Collect robot status data, including SOC data, current position node data, walking speed data, task queue data, and battery capacity data. Step S103: Collect tower passage map data, which includes node set data, edge set data, edge passage time data, and edge capacity data; Step S104: Collect charging dock status data, which includes charging dock identification, maximum power data, temperature data, contact resistance data, and availability status data. Step S105: Collect available charging power window data, which includes window start and end time data and corresponding available power upper limit data.

3. The tower bolt robot charging scheduling method as described in claim 2, characterized in that, Step S2 includes the following sub-steps: Step S201: Perform outlier removal and missing value interpolation on the tower bolt operation data, robot status data, and charging base status data to obtain cleaning data; Step S202: Synchronize the cleaned data according to a unified timestamp to obtain synchronized data; Step S203: Map the current location node data in the synchronization data to the passage map data, and unify the units of the edge passage time data to obtain the mapped data; Step S204: Merge the mapping data and power window data by timestamp to form the aligned scheduling dataset.

4. The tower bolt robot charging scheduling method as described in claim 3, characterized in that, Step S3 includes the following sub-steps: Step S301: Calculate the mechanical work data of the bolt group operation based on the torque sampling data and the angle sampling data. The mechanical work data is the result obtained by curve integration of the torque sampling data and the angle sampling data. Step S302: Calculate the task operation energy consumption prediction data based on the operation machinery power data, operation start and end time data and operation priority data, and output the task operation energy consumption prediction result; Step S303: Calculate the predicted path travel time based on the tower passage map data, current location node data, and walking speed data; Step S304: Calculate path waiting time prediction data based on the path travel time prediction data and the edge capacity data in the tower passage diagram data, and synthesize the path travel time prediction data with the path travel time prediction data to obtain path passage energy consumption prediction data.

5. The tower bolt robot charging scheduling method as described in claim 4, characterized in that, Step S4 includes the following sub-steps: Step S401: Based on the path passage energy consumption prediction data, solve the minimum energy consumption data for the robot to retreat from the current position node to the corresponding node of the nearest charging station on the tower passage map data. Step S402: Convert the minimum energy consumption data and battery capacity data for pullback into pullback power demand data; Step S403: Calculate congestion risk correction data based on edge capacity data and path waiting time prediction data; Step S404: Overlay the withdrawn power demand data with the congestion risk correction data to generate a safe power lower limit data, and write the safe power lower limit data into the aligned scheduling dataset.

6. The tower bolt robot charging scheduling method as described in claim 5, characterized in that, Step S5 includes the following sub-steps: Step S501: parse the power window data and the maximum power data of the charging dock to generate the allocatable charging power data of each charging dock within the window; Step S502: Based on the allocable charging power data, charge time slot data is generated by dividing it into segments according to a preset granularity. The charge time slot data includes charging dock identifier, time slot start and end time data, and time slot power data. Step S503: Based on the safety power limit data, filter the robots that need to be recharged, match charging time slot data for each robot, and generate charging reservation token data; Step S504: Send the charging reservation token data to the robot. The charging reservation token data includes arrival time window data, charging dock identifier, and timeout invalidation rule data.

7. The tower bolt robot charging scheduling method as described in claim 6, characterized in that, Step S6 includes the following sub-steps: Step S601: Using the task operation energy consumption prediction results, path passage energy consumption prediction data, charging time slot data and safe power limit data as inputs, construct rolling scheduling target data; Step S602: Under the conditions of satisfying the edge capacity data constraints, the charging dock available state data constraints, and the safe power lower limit data constraints, the charging scheduling command is obtained by solving. Step S603: The charging scheduling instruction and the charging reservation token data are merged to generate execution instruction data and issued. The execution instruction data includes task execution order data, path node sequence data and seated charging power data.

8. The tower bolt robot charging scheduling method as described in claim 7, characterized in that, Step S6 further includes step S604: Receive execution feedback data, which includes actual operation energy consumption data, actual path time data, charging dock temperature data, and contact resistance data; Based on the execution feedback data, the energy consumption deviation data and time slot mismatch data are calculated. When the energy consumption deviation data or time slot mismatch data exceeds the preset threshold, the corresponding charging reservation token data is revoked and the charging time slot data is rematched to generate updated charging reservation token data and updated charging scheduling instructions.

9. The tower bolt robot charging scheduling method as described in claim 8, characterized in that, The update of the alignment scheduling dataset includes: Replace the corresponding records in the task energy consumption prediction data with the actual operation energy consumption data, and correct the path waiting time prediction data with the actual path time data. Write the charging dock temperature data and contact resistance data into the charging dock status data and update the charging dock available status data; Based on the updated aligned scheduling dataset, the task operation energy consumption prediction data, safe power limit data, and charging time slot data are recalculated in a rolling manner to form the charging scheduling instructions for the next cycle.

10. A tower bolt robot charging scheduling system, applied in a tower bolt robot charging scheduling method as described in any one of claims 1-9, characterized in that, It includes a data acquisition module, a data alignment module, an energy consumption prediction module, an energy consumption calculation module, a charging reservation module, and a charging execution module; The data acquisition module is used to collect tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data; The data alignment module is used to clean, synchronize, and map the tower bolt operation data, robot status data, tower passage diagram data, charging base status data, and available charging power window data to form an aligned scheduling dataset. The energy consumption prediction module is used to calculate task operation energy consumption prediction data and path travel energy consumption prediction data based on the aligned scheduling dataset, and output the task energy consumption prediction result. The energy consumption calculation module is used to calculate the minimum energy consumption data for rollback based on the task energy consumption prediction results and generate the safe power consumption lower limit data. The charging reservation module is used to generate charging time slot data and issue charging reservation token data based on the safe power limit data and the available charging power window data. The charging execution module is used to generate charging scheduling instructions based on charging reservation token data and task energy consumption prediction results, and send them to the robot for execution, and send back execution feedback data to update the alignment scheduling dataset.

Images