CN-122022341-A - Intelligent scheduling system and scheduling method for engineering equipment

CN122022341ACN 122022341 ACN122022341 ACN 122022341ACN-122022341-A

Abstract

The invention provides an intelligent scheduling system and scheduling method for engineering equipment, and belongs to the technical field of intelligent control of engineering equipment. The method comprises the steps that a multi-source perception terminal reads and uploads standard chemical engineering equipment data to an edge node through a heterogeneous adaptation module, a cloud end utilizes unmanned aerial vehicle point cloud to construct three-dimensional terrain and digital twin scenes, virtual and real calibration is completed through an iterative nearest point algorithm, an intelligent task distribution system decomposes tasks into atomization operation units, an improved genetic algorithm is adopted to combine discrete event simulation screening optimal scheduling scheme, the edge decision node conducts conflict resolution and energy consumption optimization cooperative control based on four-dimensional space-time grids, and when abnormality is detected, D is triggered to be utilized And (5) carrying out path re-planning by an emergency response mechanism of the Lite algorithm. The method solves the problems of difficult compatibility, scheduling stiffness and multi-machine conflict of heterogeneous equipment, and realizes real-time accurate scheduling and safe closed-loop control of engineering equipment clusters.

Inventors

XIAO HAO
DONG QIFENG
XIA HAO
ZHANG YIPENG
Cheng Xuecong
YANG JUNYA
YU GUO
Luo shuang
LI GANG

Assignees

中交第二航务工程局有限公司

Dates

Publication Date: 20260512
Application Date: 20260130

Claims (10)

1. An intelligent scheduling method for engineering equipment is characterized by comprising the following steps: S1, an end side multi-source sensing terminal reads CAN bus and sensor data of engineering equipment through a heterogeneous adaptation module, data standardization is completed through a protocol analysis algorithm, and engineering equipment operation data are collected in real time and uploaded to an edge decision node; S2, the cloud scheduling center utilizes operation area point cloud data acquired by the unmanned aerial vehicle, generates a three-dimensional terrain model through a statistical outlier removal filtering and Delaunay triangulation grid construction algorithm, constructs a digital twin scene by combining engineering equipment parameters, and completes virtual-real mapping calibration through an iterative nearest point algorithm; S3, the cloud intelligent task distribution system decomposes the total construction task into discrete atomization operation units according to rated operation capacity of engineering equipment, an improved genetic algorithm with a self-adaptive cross variation mechanism is adopted to generate candidate scheduling schemes, the schemes are input into a digital twin scene, the running state of each equipment is deduced through discrete event simulation, finishing time and energy consumption data are recorded, and an optimal scheduling scheme is screened and issued to an edge decision node; S4, the edge decision node receives an optimal scheduling scheme, establishes a space-time occupation table through a conflict resolution module, and performs multi-machine cooperative control on engineering equipment by combining load mapping logic of an energy consumption optimization unit; And S5, triggering an emergency response mechanism when the end-side multi-source perception terminal detects that abnormal data of the working environment or engineering equipment exceeds a safety threshold, and carrying out global path re-planning by a cloud dispatching center through a dynamic path planning algorithm, and updating a control instruction through an edge decision node.
2. The intelligent scheduling method of engineering equipment according to claim 1, wherein in the step S1, a heterogeneous adaptation module presets a communication protocol library of multi-brand engineering equipment, the communication protocol library comprises CAN 2.0 and Modbus RTU protocols, equipment types are identified through message header identification, nonstandard CAN messages or Modbus data are converted into unified JSON format data through calling corresponding decoding rules, a state sensing unit collects data at 20Hz frequency, the data comprise GNSS position data, hydraulic system pressure data and triaxial vibration data of the engineering equipment, the GNSS position data are smoothed through a Kalman filtering algorithm, and the smoothed GNSS position data are transmitted to an edge decision node through a 5G C-V2X network.
3. The intelligent scheduling method of engineering equipment according to claim 1, wherein in step S2: S21, carrying a laser radar on an unmanned aerial vehicle to scan an operation area along a preset route with the height of 5-10m, wherein the scanning frequency is 10Hz, acquiring original point cloud data, and removing noise points by using a statistical outlier removal algorithm; s22, converting the processed point cloud data into an irregular triangular mesh model by adopting a Delaunay triangulation algorithm, extracting elevation features and gradient features of the terrain, and generating a passable regional navigation grid; S23, mapping physical dimensions, kinematic parameters and material distribution properties of engineering equipment into a three-dimensional virtual scene, and constructing a multi-level virtual environment comprising a terrain layer, an equipment layer and a material layer; S24, obtaining actual measurement GNSS coordinates of more than 3 fixed reference points on site, calculating a rotation translation matrix of corresponding feature points and the actual measurement coordinates in the virtual scene by using an iterative nearest point algorithm, and performing rigid transformation calibration on a coordinate system of the three-dimensional virtual scene.
4. The intelligent scheduling method for engineering equipment according to claim 1, wherein in the step S3, the specific method for decomposing the construction task into the atomized operation units is that the total volume V of materials in the task area and the average bucket capacity of excavating equipment participating in the operation are obtained Calculating theoretical operation times Wherein Is an upward rounding function; the atomic task allocation rule is that according to the real-time load rate and the working efficiency of each engineering device, a polling allocation combined load balancing strategy is adopted to allocate N atomic task units comprising four action sequences of digging, loading, transporting and unloading to each device, so that the atomic task sequence of a single device is ensured to be continuous and free from idle intervals.
5. The intelligent scheduling method of engineering equipment according to claim 1, wherein in step S3, the step of generating the candidate scheduling scheme by adopting the improved genetic algorithm comprises the following steps: S31, initializing a population by adopting a real number coding mode based on a task sequence, setting the population scale to be 50-100, randomly generating a task allocation sequence, and ensuring that the total number of tasks of each chromosome is equal to the total number of atomic task units; S32, defining a fitness function, wherein the specific expression of the fitness function is as follows Wherein , 、、 The weights of time, energy consumption and load variance are respectively determined according to the construction priority by an analytic hierarchy process, In order to achieve a maximum finishing time, As a result of the total energy consumption, Variance for device load; S33, performing evolution operation, wherein in the evolution process, according to the current iteration number k and the maximum iteration number Dynamically adjusting crossover rate Mutation rate The adjustment formula is 、 Wherein For an initial crossover ratio of 0.8-0.9, To terminate the crossover ratio 0.4-0.5, The initial mutation rate is 0.05-0.1, The termination mutation rate is 0.01-0.02; S34, simulation screening, namely starting a virtual clock in a digital twin scene for the generated candidate scheduling scheme, setting time parameters of each action by taking an action sequence of an atomic task as an event unit and combining rated parameters of engineering equipment, scheduling the operation of the simulation equipment through the event queue, recording finishing time and energy consumption data at the end of simulation as fitness evaluation basis, and enabling evolution termination conditions to be reached according to iteration times Or continuously 10 generations of maximum value of fitness function is not lifted, and a scheduling scheme with optimal fitness is screened out.
6. The intelligent scheduling method of engineering equipment according to claim 1, wherein in step S4, the specific method for controlling by the conflict resolution module is as follows: S41, constructing four-dimensional space-time grid of operation area Wherein i, j, k, l are grid indexes, the grid resolution of the space dimension (x, y, z) is set to be 0.5m x 0.5m, the time slice of the time dimension t is set to be 0.1s, and the planning path of each engineering equipment is mapped into a series of space-time voxel occupation requests; s42, conflict detection, safety threshold According to the maximum width of engineering equipment and the safety margin of 0.3m, if two pieces of equipment are detected to be in the same time slice The internal requests occupy the same space element Or the spatial distance is less than the safety threshold Judging that the paths conflict; S43, conflict resolution, namely calculating priority weights of all conflict equipment tasks Wherein In order to achieve a degree of task urgency, For the importance of the task, Keeping the high priority equipment path unchanged, applying a time lag amount to the low priority equipment And d is the distance between the conflict point and the current position of the low-priority equipment, v is the rated running speed of the equipment, and the time space occupation request is updated until the conflict disappears, so that a conflict-free control instruction is generated and issued.
7. The intelligent scheduling method of engineering equipment according to claim 1, wherein in step S4, the control method of the energy consumption optimizing unit is as follows: s44, acquiring pump outlet pressure data P and flow data Q of the hydraulic system of the excavating equipment in real time, wherein the pump outlet pressure data P and the flow data Q are obtained through the formula Reversely estimating the current earthwork hardness value H, wherein k is a correction coefficient, sand k=0.8, clay k=1.2, Is the operation angular velocity; S45, acquiring the transportation distance L and the road gradient of the current task ; S46, inquiring a preset three-dimensional MAP (MAP) chart of hardness, distance of transportation and power, wherein the horizontal axis of the three-dimensional MAP chart is earthwork hardness H, the vertical axis of the three-dimensional MAP chart is transportation distance L, the vertical axis of the three-dimensional MAP chart is engine power P, supplementing non-preset working condition parameters through linear interpolation, determining an optimal fuel economy working point of the engine under the current working condition, outputting a corresponding target rotating speed instruction to an engine electronic control unit, adjusting the fuel injection quantity to match a target power output curve through a PID (proportion integration differentiation) control algorithm, and presetting a target rotating speed instruction corresponding fuel injection quantity adjusting range according to the engine parameters.
8. The intelligent scheduling method of engineering equipment according to claim 1, wherein in step S5, the multi-source sensing terminal comprises a triaxial acceleration sensor, and the emergency response mechanism comprises the following steps: S51, performing Fast Fourier Transform (FFT) on time domain vibration signals acquired by a triaxial acceleration sensor, and extracting a frequency domain characteristic value; S52, presetting frequency domain characteristic thresholds of different fault types, wherein bearing abrasion corresponds to a frequency band amplitude threshold of 100-200Hz and is 5g-8 g, hydraulic leakage corresponds to a frequency band amplitude threshold of 50-100Hz and is 3g-5g, calibrating through historical fault data, and if the amplitude of a certain frequency band exceeds the corresponding preset fault characteristic threshold, immediately generating and uploading an alarm signal by a local rule engine; S53, after the cloud dynamic priority engine receives the alarm, locking related equipment around the fault equipment, taking the fault equipment as a center, judging all engineering equipment in a round area with the radius of 5-8m as related equipment, suspending an original task, and marking the position of the fault equipment as a dynamic obstacle; S54, adopt D Lite dynamic path planning algorithm, heuristic function is set as Wherein As the current location is to be determined, For a safe evacuation point, marking the position of the fault equipment as a high-cost area, synchronously updating a navigable area navigation grid, searching an avoidance path in the updated cost map, generating an emergency control instruction sequence containing a steering angle and a speed, and re-planning the maximum turning angle of the path to be not more than 30 degrees.
9. An intelligent scheduling system for engineering equipment is characterized in that an end, side and cloud three-level cooperative architecture is adopted, and the intelligent scheduling system is matched with the intelligent scheduling method for engineering equipment according to any one of claims 1-8, and comprises the following components: The multi-source sensing terminal is arranged on engineering equipment, hardware comprises a CAN bus adapter, a GNSS positioning module, a triaxial acceleration sensor and a hydraulic sensor, software comprises a protocol analysis plug-in unit and a Kalman filtering program, comprises a heterogeneous adaptation module and a state sensing unit and is used for executing protocol analysis, data filtering, standardized uploading and anomaly detection of S1 step and S51-S52 step; The edge decision node is deployed on an operation site edge server, the edge server is an industrial-grade edge server, and software comprises a real-time operating system, a conflict resolution algorithm plug-in and an energy consumption control program, and comprises a local rule engine, a conflict resolution module and an energy consumption optimization unit, and the edge decision node is used for executing space-time conflict detection of S4 steps, S41-S43 steps and S44-S46 steps and issuing engine power dynamic regulation and control instructions; The cloud scheduling center is deployed on the cloud server cluster, and the software comprises a digital twin platform, a genetic algorithm engine and a path planning module, and comprises a digital twin platform, an intelligent task distribution system and a safety supervision module, wherein the digital twin platform is used for executing point cloud data processing, scene construction, genetic algorithm task distribution, global path re-planning and emergency process monitoring in the steps S2-S3, S31-S34 and S53-S54; The multi-source sensing terminal is communicated with the edge decision node through a 5G C-V2X network, the edge decision node is communicated with the cloud scheduling center through an optical fiber Ethernet, the data interaction format is unified as JSON, and the transmission frequency is 20Hz.
10. The intelligent scheduling system for engineering equipment based on digital twinning and improved genetic algorithm according to claim 9, wherein the intelligent task allocation system is configured to perform the construction task decomposition and improved genetic algorithm steps; the digital twin platform is configured to perform the point cloud processing and virtual-real calibration steps; the conflict resolution module is configured to perform a four-dimensional space-time grid conflict detection step; the energy consumption optimizing unit is configured to execute the energy consumption adjustment control step; The local rule engine is matched with the multi-source perception terminal and is configured to execute an abnormality detection and alarm step; the safety supervision module is configured to receive the alarm signal, monitor the emergency response process and synchronously update the device state in the digital twin scene.

Description

Intelligent scheduling system and scheduling method for engineering equipment Technical Field The invention relates to the field of intelligent scheduling of engineering equipment, in particular to an intelligent scheduling system and an intelligent scheduling method of engineering equipment. Background In large-scale construction scenes such as mines, ports, municipal works and the like, the scheduling efficiency of engineering equipment such as an excavator, a loader, a dump truck and the like directly determines the construction progress, cost and safety. With the development of industrial internet and intelligent manufacturing technology, intelligent scheduling of engineering equipment has become a necessary trend of industry development. However, the current engineering equipment scheduling field still faces a number of technical bottlenecks. Firstly, the compatibility problem of heterogeneous equipment is solved, the communication protocols of engineering equipment with different brands and different models are not uniform, and a plurality of protocols such as a CAN bus, a Modbus, an Ethernet and the like coexist, so that the data acquisition is difficult, and a uniform control network is difficult to form. Secondly, task allocation is often static and stiff, and the traditional scheduling method lacks dynamic response capability to equipment real-time working conditions and operation environment changes, so that global optimization is difficult to realize. Furthermore, when multiple machines cooperate to operate, especially when large-scale heterogeneous equipment clusters are involved, path planning is often self-administration, and a unified space-time cooperation mechanism is lacked, so that cross operation conflicts frequently occur, and potential safety hazards are large. In the prior art, the prior art with the patent number of CN118246310A discloses a digital twin modeling method for equipment remote intelligent monitoring and operation and maintenance, and the technology realizes comprehensive monitoring and association analysis on equipment working conditions, process logistics and processing quality by constructing an industrial flow data acquisition system architecture with layered design and utilizing a multi-mode geometric model construction method and a dynamic event-state knowledge graph. However, this technique is mainly focused on digital twin modeling at the "monitoring" and "operation and maintenance" level, focusing on how to accurately map the state of the physical world for presentation and analysis, and lacks an "active scheduling" strategy for large-scale heterogeneous equipment clusters. Although the data acquisition architecture is established, the deep analysis and unified standardization adaptation of the underlying heterogeneous protocols (such as non-standard CAN messages) of different brands of engineering equipment are not elaborated, and the dynamic task allocation and multi-machine collaborative path conflict resolution mechanisms based on intelligent optimization algorithms such as genetic algorithms are not involved, so that the method is difficult to be directly applied to complex field construction scheduling scenes. In addition, the prior art with the patent number of CN120125180A discloses a digital engineering dispatching optimization method and device based on a discrete event simulation technology, wherein the technology is used for simulating a construction task execution process by constructing a digital twin model of project dispatching and combining a multi-objective optimization algorithm (such as a genetic algorithm and a particle swarm algorithm) to carry out task dispatching optimization and construction period prejudgment. In the scheme, although a simulation and optimization algorithm is introduced to solve the scheduling problem, in the practical floor application, the requirements of hardware deployment and instantaneity of an end-side-cloud three-level collaborative architecture are not considered enough. Particularly in terms of real-time decision capability of the edge side of a job site, the technology lacks a specific implementation means for millisecond path conflict detection and resolution, and does not establish a refined occupation model based on four-dimensional space-time grids. Meanwhile, the technology mainly focuses on macroscopic resource allocation in the aspect of energy consumption optimization, lacks a refined closed-loop control logic for reversely adjusting the power output of the engine of the equipment based on physical properties (such as earthwork hardness) of an operation object, and cannot realize the extreme energy conservation of engineering equipment in a microscopic level. In summary, the prior art is either biased to static monitoring modeling or to macroscopic task scheduling simulation, and lacks an intelligent scheduling system which can deeply fuse heterogeneous equipment bottom data, has end-to-side cloud coo