Parallel control method, system and equipment for movable umbrella-shaped crossing frame and storage medium

Abstract

A parallel control method for a mobile umbrella-shaped spanning frame comprises the following steps: constructing a construction scene digital twin model based on a real scene to be constructed, constructing each mobile spanning frame digital twin model, and constructing a parallel control model for connecting the mobile spanning frame digital twin model and an actually constructed mobile spanning frame system; constructing a knowledge graph of the movable umbrella-shaped crossing frame and a construction operation scene based on a construction line design construction scheme and construction safety specifications and standards; based on the digital twinborn model and the knowledge graph, constructing a construction plan of the mobile umbrella-shaped spanning frame, and performing construction deduction in the digital twinborn model to obtain a construction planning scheme; and on-site construction parallel control is carried out by taking a construction planning scheme as a target on the basis of a parallel control model for connecting the mobile spanning frame digital twinborn model and an actually constructed mobile spanning frame system. Safety risks are avoided in advance in the construction process through digital intelligent planning, it is guaranteed that the construction scheme is reasonably developed, and follow-up construction is orderly guided.

Description

Technical Field

[0001] The present invention relates to a parallel control method, system, device and storage medium for a mobile umbrella-shaped spanning frame, which are specifically suitable for realizing real-time risk management and intelligent control during the construction process of a line spanning scene. Background Art

[0002] Mobile crossing frames are used for mechanized construction in cross-over construction. Taking into account the complex environment of the line construction site, mechanized construction is restricted by natural conditions and other factors, and there is a lack of detailed and scientific construction plan planning and deduction tools. At present, there is still a lack of research on construction planning and dynamic deduction in the construction scenario of mobile crossing frames. At the same time, with the gradual deepening of the digitalization of construction site equipment, the processing of large amounts of data and the management and control of complex systems have become urgent problems that affect construction safety and efficiency. Managers are also eager to have a more efficient digital management and control method.

[0003] In response to these problems, it is imperative to propose a new generation of multi-data-driven intelligent management framework to meet future challenges. The concept of parallel system and digital twin provides a new solution paradigm for solving the above problems. Parallel system can achieve better management and control of complex systems. It can generate a large number of scenes based on artificial systems, and complete the planning and deduction of construction scenes, analyze the optimal control scheme, and realize adaptive optimization control. Digital twins make full use of sensing technology, big data technology, multi-disciplinary, multi-physical quantity, and multi-scale simulation modeling and analysis, and use model-driven, data-driven, and joint-driven modeling methods to establish virtual entities synchronized with the entire life cycle of physical entities in the digital world, simulate and predict the characteristics of physical entities and their changing laws. The parallel system construction of the present invention draws on the modeling method of digital twins to achieve the highest accuracy and maximum fidelity model as much as possible, but at the same time follows the basic concept of parallel systems, overcoming the shortcomings that digital twins can only predict future changes based on the actual data of physical entities, cannot evaluate the system performance of multiple schemes and multiple parameters, and are prone to fall into local optimality. The implementation of the parallel system has expanded the scope and boundaries of the information space for mechanized construction of transmission lines, better simulated complex systems, provided a more comprehensive optimization control solution, and established a new solution for traditional power grid construction methods in the digital age.

[0004] Based on the above, in order to achieve the goal of improving the construction safety and efficiency of mobile umbrella-shaped crossing frames, how to construct a three-dimensional digital model corresponding to the mobile umbrella-shaped crossing frame, how to construct a knowledge map of the crossing frame construction and complete the digital planning and deduction of the entire construction process, and how to realize autonomous parallel control of the crossing frame through mechanical optimal control theory based on the planning and deduction results are the key issues in establishing a parallel control system for mobile umbrella-shaped crossing frames. Summary of the invention

[0005] The purpose of the present invention is to overcome the problems existing in the prior art of complex line construction site environment, mechanized construction being restricted by natural conditions and other factors, and low on-site construction efficiency, and to provide a mobile umbrella-shaped spanning frame parallel control method, system, equipment and storage medium that automatically plans construction plans and realizes parallel control during construction.

[0006] To achieve the above objectives, the technical solution of the present invention is:

[0007] In a first aspect, the present invention provides a method for parallel control of a mobile umbrella-shaped spanning frame, comprising the following steps:

[0008] Build a digital twin model of the construction scene based on the real construction scene, build digital twin models of each mobile crossing frame based on different crossing frame models, and build a parallel control model connecting the digital twin model of the mobile crossing frame and the mobile crossing frame system under actual construction;

[0009] Based on the design and construction plan of the line to be constructed, the actual construction plan of the historical construction line, and the current construction safety specifications and standards, a knowledge graph of mobile umbrella-shaped spanning frames and construction operation scenarios is constructed;

[0010] Based on the digital twin model and knowledge graph, a construction plan for a mobile umbrella-shaped spanning frame is constructed, and construction simulation is performed in the digital twin model to obtain a construction plan.

[0011] Based on the parallel control model connecting the digital twin model of the mobile crossing frame and the actual construction of the mobile crossing frame system, parallel control of on-site construction is carried out with the construction planning scheme as the target.

[0012] A digital twin model of the construction scene is constructed based on the real construction scene, and the relevant parameter information of the line to be constructed is imported into the model, including the distribution coordinate information of each tower, tower model, tower number and layout and installation design information of the transmission line; at the same time, the domestic road network information based on the map is synchronized in proportion on the bottom layer of the model, so that the coordinates of the digital twin scene correspond to the coordinates of the map one by one, and a digital twin model of the construction scene is obtained;

[0013] Based on different types of crossing frames, digital twin models of mobile crossing frames are constructed. A three-dimensional geometric model is obtained through parametric design. The physical, behavioral and rule models of each subsystem are defined according to the actual operation mode of the mobile crossing frame to form a high-fidelity mapping of the physical entity.

[0014] Then, digital twin models are built for the construction equipment and materials that are used in conjunction with the spanning frame.

[0015] The constructed digital twin model of the mobile spanning frame and the digital twin models of the construction equipment and materials can all be arranged in the digital twin model of the construction scene according to the construction plan;

[0016] A parallel control model connecting the digital twin model of the mobile crossing frame and the mobile crossing frame system under actual construction is constructed, so that the control signal of the digital twin model of the mobile crossing frame can synchronously control the mobile crossing frame system under actual construction. At the same time, the mobile crossing frame system under actual construction will feed back the received mobile crossing frame sensor signal to the digital twin model of the mobile crossing frame to assist the feedback adjustment of the parallel control model.

[0017] The method for constructing a parallel control model connecting the digital twin model of the mobile crossing frame and the mobile crossing frame system in actual construction is as follows:

[0018] The parallel control problem of the straddle rack is essentially the target tracking problem of the target speed and position of the straddle rack. The time of the target tracking problem is discretized. For a certain moment, the target expected state of the straddle rack at time t+1 is:

[0019] x d [t+1]=A d x d [t]

[0020] Among them, A d is the target transfer matrix, x d [t] is the expected state of the spanning rack at time t, x d [t+1] is the expected state of the spanning rack at time t+1;

[0021] Construct the spanning frame state equation:

[0022] x[t+1]=Ax[t]+Bu[t]

[0023] Among them, x[t+1] is the state target of the spanning rack at time t+1, A is the state matrix of the spanning rack, B is the input control matrix of the spanning rack, and u[t] is the input control vector of the spanning rack system at time t;

[0024] Since the passive target velocity and posture of the straddle frame are both constant vectors, A d =I; Definition error:

[0025] e[t]=x[t]-x d [t]

[0026] Where e[t] is the control error of the spanning frame at time t;

[0027] At this point, the target tracking problem is transformed into a type of error adjustment problem; integrating the above formula:

[0028]

[0029] At this time, the augmented form space state equation is obtained: a [t+1]=A a x a [t]+B a u[t]

[0030] where x a [t+1] is the state vector of the spanning rack at time t+1 after merging, A a is the state matrix across the frame after merging, x a [t] is the state vector of the spanning rack at time t after merging, B a is the input control matrix of the merged spanning rack; then:

[0031]

[0032] Among them, C a for [I -I];

[0033] Relying on parallel control theory, the cost function J can be constructed:

[0034]

[0035] Among them, S is a symmetric positive definite matrix, which is usually used to weigh the importance of the state error e[t] at time t, Q is the state cost weighting matrix, R is the control cost weighting matrix, and N represents the previous N moments;

[0036] This cost function is the cost function construction in the tracking problem: represents the error between the current state and the expected state at the current time; e[t] T Qe[t] represents the cost or energy required to reduce the error accumulated in the past; u[t] T Ru[t] is the energy required for the system to reach the next state at this moment; substituting e[t] into J, we get:

[0037]

[0038] Let S a =C a T SC a , Q a =C a T QC a , the cost function is:

[0039]

[0040] At this time, the cost function and the augmented form space state equation conform to the linear quadratic standard form; through the above process, a type of error tracking problem is successfully converted into a type of regulation problem;

[0041] The linear quadratic controller uses state feedback u[t] = -Kx a [t] to minimize the cost function, we get:

[0042]

[0043] Assume a constant matrix P that satisfies:

[0044]

[0045] Take K = R -1 B a T P, then A a T P+PA a +Q a -PB a R -1 B a T P=0This equation is the Riccati equation;

[0046] The calculation process of K: First, the state cost weighted matrix Q is determined by the Bryson method a and the control cost weight matrix R, Q a is a semi-positive definite matrix, and R is a positive definite matrix; then, solve the Riccati equation to obtain the matrix P; finally, calculate K = R -1 B a T P.

[0047] Based on the design and construction plan of the line to be constructed, the actual construction plan of the historical construction line, and the current construction safety specifications and standards, a knowledge graph of mobile umbrella-shaped spanning frames and construction operation scenarios is constructed;

[0048] In the process of knowledge graph construction, construction organization requirements, construction process specifications, construction organization structure, and resource allocation are the main clues;

[0049] Knowledge update: Based on the newly generated or modified construction plan of the mobile umbrella-shaped spanning frame construction planning and simulation model, the knowledge graph is updated and stored. The newly added construction plan needs to be added to the original knowledge graph after knowledge fusion. At the same time, it should be considered whether the same entity or relationship already exists in the existing knowledge graph for the newly added entity or relationship. If so, deduplication should be performed; whether there are invalid entities or relationships after the knowledge update, if so, they should be eliminated;

[0050] For modifications to construction safety specifications and standards, the knowledge graph is updated and stored. For data where relevant provisions in new construction safety specifications and standards conflict with past provisions, the knowledge graph generates a comparison list and determines it manually.

[0051] Based on the digital twin model and knowledge graph, a construction plan for a mobile umbrella-shaped spanning frame is constructed, and construction deduction is performed in the digital twin model to obtain a construction plan: the relevant information parameters in the digital twin model of the construction scene are input into the knowledge graph. The knowledge graph generates an initial construction plan based on the construction points of the construction plan for the line to be constructed. In the generation process, the actual construction plans of similar construction points in the historical construction line and the current construction safety specifications and standards need to be considered;

[0052] The initial construction plan generated by the knowledge graph is adjusted based on the actual situation. The knowledge graph corrects the number or parameters of related entities based on the adjustment of the construction plan to improve the construction plan.

[0053] The improved construction plan is input into the digital twin model. The digital twin model calls the spanning frame, construction equipment, and construction materials in the construction plan and arranges them in the digital twin model of the construction scene. The spanning frame, construction equipment, and construction materials are manually arranged and adjusted, and then the construction simulation of the spanning frame is started. If the simulation is successful, the construction plan is obtained. If the simulation fails, the cause of the failure is analyzed, and the parameters are adjusted before re-simulation.

[0054] During construction simulation:

[0055] Carry out safety verification calculations for the post-construction planning to confirm that the construction load-bearing capacity meets the construction standard requirements;

[0056] Based on the high-precision three-dimensional model of the mobile crossing frame operation scene, the deduction function is used to realize the dynamic simulation of the construction plan of the mobile crossing frame of the transmission and transformation line. According to the setting of the construction process, the construction content of each stage in the construction process is dynamically simulated in a time sequence manner; in the deduction process, the spanning frame deployment angle, crane working range, arm length, angle and other parameters collected by the sensor are used to verify the safety of the spanning frame and the crane when operating at the crossing point during the construction process. At the same time, the safe distance between the crane and the live body during the operation, and the safe distance between the capping net and the live wire and the ground wire are calculated; if it is not within the safe distance, the alarm will be immediately shut down, and the deduction will be repeated after manual adjustment.

[0057] The construction planning and construction simulation are both carried out in the digital twin model. The digital twin model displays the digital twin scene with planned construction in the form of a background display, and the generation results and modifications of the construction planning are displayed in the form of a catalog list. The construction simulation is displayed in the digital twin scene in the form of a proportional restoration of the digital twin span, the construction equipment and materials supporting the span.

[0058] In a second aspect, the present invention provides a mobile umbrella-type spanning frame parallel control system, the system is used to execute the aforementioned mobile umbrella-type spanning frame parallel control method, specifically comprising: a digital twin, a parallel control model construction module, a knowledge graph construction module, a construction planning and deduction model construction module, and a parallel control implementation module;

[0059] Digital twin and parallel control model construction module: used to build a digital twin model of the construction scene based on the real construction scene, build digital twin models of each mobile crossing frame based on different crossing frame models, and build a parallel control model connecting the digital twin model of the mobile crossing frame and the mobile crossing frame system under actual construction;

[0060] Knowledge graph construction module: used to construct a knowledge graph of mobile umbrella-shaped spanning frames and construction operation scenarios based on the design and construction plan of the line to be constructed, the actual construction plan of the historical construction line, and the current construction safety specifications and standards;

[0061] Construction planning and deduction model building module: used to build the construction planning of the mobile umbrella-shaped spanning frame based on the digital twin model and knowledge graph, and to perform construction deduction in the digital twin model to obtain the construction planning scheme;

[0062] Parallel control implementation module: It is used to perform parallel control of on-site construction based on the parallel control model connecting the digital twin model of the mobile crossing frame and the actual construction of the mobile crossing frame system, with the construction planning scheme as the target.

[0063] In a third aspect, the present invention provides a mobile umbrella-type spanning frame parallel control device, comprising a memory and a processor, wherein the memory is used to store a computer program code and transmit the computer program code to the processor;

[0064] The processor is used to execute the aforementioned parallel control method of the mobile umbrella-shaped spanning frame according to the instructions in the computer program code.

[0065] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the aforementioned method for parallel control of a mobile umbrella-shaped spanning frame.

[0066] Compared with the prior art, the present invention has the following beneficial effects:

[0067] 1. In the parallel control method of a mobile umbrella-shaped spanning frame of the present invention, when establishing the system, it is allowed to control the actions of the system according to the artificial social content such as human entity activities and habitual thinking. By utilizing various information technologies, "people" are truly included in the engineering system and management process. The theoretical advantages of parallel systems in computational experiments are highlighted. Based on high-precision three-dimensional digital models and knowledge graph technology, the mobile spanning frame construction scheme association analysis is constructed to form a knowledge base and standardized adaptation principles, which plays an auxiliary decision-making role in the selection of process methods in construction, and forms a set of mechanized construction planning and deduction system methods. This method not only provides detailed operation guidance for on-site construction, but is also an important technical component of the digital system. In the construction planning stage, the entire construction process is planned using knowledge graphs and standardized adaptation principles. At the same time, based on technical frameworks such as parallel systems and digital twins, the deduction technology of the construction scheme of the mobile spanning frame is studied. The construction scene generation of the mobile spanning frame is completed in the artificial system, and the optimization scheme is analyzed through multiple trial and error experiments to achieve full process information interaction, full stage real-time feedback, and full cycle optimization control. Through digital planning and simulation, safety risks can be avoided in advance during the construction process, and preventive and parallel control measures can be taken to ensure that the construction plan is reasonably implemented on site and to guide subsequent construction in an orderly manner.

[0068] 2. In the parallel control method of a mobile umbrella-shaped spanning frame of the present invention, the reasonable sensor arrangement scheme and data collection of the spanning frame, the digital planning of the spanning frame construction scene and the construction deduction based on physical laws are used to realize the autonomous generation of the spanning frame construction plan, select the optimal plan based on the trial and error experiment idea and give the corresponding parallel control instructions, reduce the danger of the construction process, and improve the intelligence and digitization of the line construction equipment. (1) The implementation of the computational experiment function in the spanning frame parallel control system can realize the generation of the whole process construction planning scheme of the spanning frame operation process according to the existing conditions and the requirements of the specification, improve the decision-making efficiency through multi-dimensional measurement and comparison of the multi-objective optimization algorithm, and provide the best plan for the construction deduction; (2) The construction deduction generates detailed construction steps according to the construction planning content in the artificial system, and the planning deduction is used as the basis for parallel control calculation, and the deduction result can be used as the target value of parallel control; (3) Parallel control can realize the simultaneous processing of multiple tasks, thereby accelerating the solution of problems and the formulation of decisions; (4) The spanning frame parallel control system provides a visual representation of the real-time equipment operation status and the surrounding environment of the spanning frame, and provides operation instructions for the autonomous optimal control of the spanning construction.

[0069] 3. A mobile umbrella-shaped spanning frame parallel control system of the present invention includes a digital twin and a parallel control model construction module: used to construct a construction scene digital twin model based on a real construction scene, construct digital twin models of each mobile spanning frame based on different spanning frame models, and construct a parallel control model connecting the digital twin model of the mobile spanning frame and the mobile spanning frame system of actual construction; a knowledge graph construction module: used to construct a knowledge graph of the mobile umbrella-shaped spanning frame and the construction operation scene based on the design construction plan of the line to be constructed and the actual construction plan of the historical construction line and the current construction safety specifications and standards; a construction planning and deduction model construction module: used to construct a mobile umbrella-shaped spanning frame construction plan based on the digital twin model and the knowledge graph, and perform construction deduction in the digital twin model to obtain a construction planning plan; a parallel control implementation module: used to perform parallel control of on-site construction based on the parallel control model connecting the digital twin model of the mobile spanning frame and the mobile spanning frame system of actual construction with the construction planning plan as the target. The system is used to implement the steps of the mobile umbrella-shaped spanning frame parallel control method provided in any of the above technical solutions. Therefore, the system also includes all the beneficial effects of the parallel control method based on the movable umbrella-shaped spanning frame provided in any of the above-mentioned technical solutions, which will not be repeated here.

[0070] 4. A mobile umbrella-type spanning frame parallel control device of the present invention includes a processor and a memory, the memory is used to store computer program codes and transmit the computer program codes to the processor, and the processor is used to execute the mobile umbrella-type spanning frame parallel control method provided in any of the above technical solutions according to the instructions in the computer program codes. Therefore, the device also includes all the beneficial effects of the mobile umbrella-type spanning frame parallel control method provided in any of the above technical solutions, which will not be repeated here.

[0071] 5. The present invention provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the steps of the parallel control method of a mobile umbrella-shaped spanning frame provided in any of the above technical solutions are implemented. Therefore, the computer program product also includes all the beneficial effects of the parallel control method of a mobile umbrella-shaped spanning frame provided in any of the above technical solutions, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a flow chart of the method of the present invention.

[0073] Figure 2 It is a system module diagram of the present invention.

[0074] Figure 3 It is a schematic diagram of the device of the present invention.

[0075] Figure 4It is the implementation roadmap of the parallel control model of the movable crossing frame in Example 1.

[0076] Figure 5 It is the design planning construction drawing in Example 1.

[0077] Figure 6 It is a schematic diagram of the adjustment of the construction plan in Example 1.

[0078] Figure 7 It is a schematic diagram of the simulation of the spanning frame in Example 1.

[0079] Figure 8 It is a schematic diagram of calculating the safe distance for construction of the crossing frame in Example 1. DETAILED DESCRIPTION

[0080] The present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0081] Embodiment 1:

[0082] See also Figure 1 , Figure 4-8 A parallel control method for a movable umbrella-shaped spanning frame comprises the following steps:

[0083] Build a digital twin model of the mobile umbrella-shaped spanning frame and the construction operation scene, and connect the digital twin model of the mobile spanning frame with the parallel control model of the mobile spanning frame system under actual construction;

[0084] A digital twin model of the construction scene is constructed based on the real construction scene, and the relevant parameter information of the line to be constructed is imported into the model, including the distribution coordinate information of each tower, tower model, tower number and layout and installation design information of the transmission line; at the same time, the domestic road network information based on the map is synchronized in proportion on the bottom layer of the model, so that the coordinates of the digital twin scene correspond to the coordinates of the map one by one, and a digital twin model of the construction scene is obtained;

[0085] Based on the existing different types of crossing frames, digital twin models of mobile crossing frames are constructed respectively. The three-dimensional geometric model is obtained through parametric design. The physical, behavioral and rule models of each subsystem are defined according to the actual operation mode of the mobile crossing frame to form a high-fidelity mapping of the physical entity.

[0086] Then, digital twin models are built for the construction equipment and materials that are used in conjunction with the spanning frame.

[0087] The constructed digital twin model of the mobile spanning frame and the digital twin models of the construction equipment and materials can all be arranged in the digital twin model of the construction scene according to the construction plan;

[0088] A parallel control model connecting the digital twin model of the mobile crossing frame and the mobile crossing frame system under actual construction is constructed, so that the control signal of the digital twin model of the mobile crossing frame can synchronously control the mobile crossing frame system under actual construction. At the same time, the mobile crossing frame system under actual construction will feed back the received mobile crossing frame sensor signal to the digital twin model of the mobile crossing frame to assist the feedback adjustment of the parallel control model.

[0089] The method of building a parallel control model is as follows:

[0090] The parallel control problem of the straddle rack is essentially the target tracking problem of the target speed and position of the straddle rack. The time of the target tracking problem is discretized. For a certain moment, the target expected state of the straddle rack at time t+1 is:

[0091] x d [t+1]=A d x d [t]

[0092] Among them A d is the target transfer matrix, x d [t] is the expected state of the spanning rack at time t, x d [t+1] is the expected state of the spanning rack at time t+1;

[0093] Construct the state equation of the spanning frame: x[t+1]=Ax[t]+Bu[t];

[0094] Among them, x[t+1] is the state target of the spanning rack at time t+1, A is the state matrix of the spanning rack, B is the input control matrix of the spanning rack, and u[t] is the input control vector of the spanning rack system at time t;

[0095] Since the passive target velocity and posture of the straddle frame are both constant vectors, A d =I; Definition error:

[0096] e[t]=x[t]-x d [t];

[0097] Where e[t] is the control error of the spanning frame at time t;

[0098] At this point, the target tracking problem is transformed into a type of error adjustment problem; integrating the above formula:

[0099]

[0100] At this time, the augmented form space state equation is obtained:

[0101] x a [t+1]=A a x a [t]+Ba u[t]

[0102] where x a [t+1] is the state vector of the spanning rack at time t+1 after merging, A a is the state matrix across the frame after merging, x a [t] is the state vector of the spanning rack at time t after merging, B a is the input control matrix of the merged spanning rack; then:

[0103]

[0104] Among them, C a for [II];

[0105] Relying on parallel control theory, the cost function J can be constructed:

[0106]

[0107] Among them, S is a symmetric positive definite matrix, which is usually used to weigh the importance of the state error e[t] at time t, Q is the state cost weighting matrix, R is the control cost weighting matrix, and N represents the previous N moments;

[0108] This cost function is the cost function construction in the tracking problem: represents the error between the current state and the expected state at the current time; e[t] T Qe[t] represents the cost or energy required to reduce the error accumulated in the past; u[t] T Ru[t] is the energy required for the system to reach the next state at this moment;

[0109] Substituting e[t] into J, we get:

[0110]

[0111] Let S a =C a T SC a , Q a =C a T QC a , the cost function is:

[0112]

[0113] At this time, the cost function and the augmented form space state equation conform to the linear quadratic standard form; through the above process, a type of error tracking problem is successfully converted into a type of regulation problem;

[0114] The goal of the linear quadratic controller is to design a controller that provides a state feedback u[t] = -Kx a [t] to minimize the cost function, we get:

[0115]

[0116] Assume a constant matrix P that satisfies:

[0117]

[0118] Take K = R -1 B a T P, then A a T P+PA a +Q a -PB a R -1 B a T P=0This equation is the Riccati equation;

[0119] The calculation process of K: First, the state cost weighted matrix Q is determined by the Bryson method a and the control cost weight matrix R, Q a is a semi-positive definite matrix, and R is a positive definite matrix; then, solve the Riccati equation to obtain the matrix P; finally, calculate K = R -1 B a T P.

[0120] Construct a knowledge graph of mobile umbrella-shaped spanning frames and construction operation scenarios; including:

[0121] Based on the design and construction plan of the line to be constructed, the actual construction plan of the historical construction line, and the current construction safety specifications and standards ("Safety Work Regulations for Electric Power Construction", "Standard Process for Power Transmission and Transformation Engineering of State Grid Corporation of China", "Q / GDW11957.2-2020 Safety Work Regulations for Electric Power Construction of State Grid Corporation of China", and "DL / T5106-2017 Construction Regulations for Crossing Power Lines"), a knowledge graph of mobile umbrella-shaped spanning frames and construction operation scenarios is constructed;

[0122] In the process of knowledge graph construction, construction organization requirements, construction process specifications, construction organization structure, and resource allocation are the main clues;

[0123] Knowledge update: Based on the newly generated or modified construction plan of the mobile umbrella-shaped spanning frame construction planning and simulation model, the knowledge graph is updated and stored. The newly added construction plan needs to be added to the original knowledge graph after knowledge fusion. At the same time, it should be considered whether the same entity or relationship already exists in the existing knowledge graph for the newly added entity or relationship. If so, deduplication should be performed; whether there are invalid entities or relationships after the knowledge update, if so, they should be eliminated;

[0124] For modifications to construction safety specifications and standards, the knowledge graph is updated and stored. For data where relevant provisions in new construction safety specifications and standards conflict with past provisions, the knowledge graph generates a comparison list and determines it manually.

[0125] Based on the digital twin model and knowledge graph, a construction plan for a mobile umbrella-shaped spanning frame is constructed, and construction simulation is performed in the digital twin model to obtain a construction plan.

[0126] The construction planning and construction simulation are both carried out in the digital twin model. The digital twin model presents the digital twin scene with planned construction in the form of a background display, and the generation results and modifications of the construction planning are displayed in the form of a catalog list. The construction simulation is displayed in the digital twin scene in the form of a proportional restoration of the digital twin spanning frame, the construction equipment and materials supporting the spanning frame.

[0127] The relevant information parameters in the digital twin model of the construction operation scene are input into the knowledge graph. The knowledge graph generates an initial construction plan based on the construction points of the construction plan for the line to be constructed. In the generation process, it is necessary to consider the actual construction plans of similar construction points in the historical construction line and the current construction safety specifications and standards;

[0128] The initial construction plan generated by the knowledge graph is adjusted based on the actual situation. The knowledge graph corrects the number or parameters of related entities based on the adjustment of the construction plan to improve the construction plan.

[0129] The improved construction plan is input into the digital twin model. The digital twin model calls the spanning frame, construction equipment, and construction materials in the construction plan and arranges them in the digital twin model of the construction scene. The spanning frame, construction equipment, and construction materials are manually arranged and adjusted, and then the construction simulation of the spanning frame is started. If the simulation is successful, the construction plan is obtained. If the simulation fails, the cause of the failure is analyzed, and the parameters are adjusted before re-simulation.

[0130] Carry out safety verification calculations for the post-construction planning to confirm that the construction load-bearing capacity meets the construction standard requirements;

[0131] Based on the high-precision three-dimensional model of the mobile crossing frame operation scene, the deduction function is used to realize the dynamic simulation of the construction plan of the mobile crossing frame of the transmission and transformation line. According to the setting of the construction process, the construction content of each stage in the construction process is dynamically simulated in a time sequence manner; in the deduction process, the spanning frame deployment angle, crane working range, arm length, angle and other parameters collected by the sensor are used to verify the safety of the spanning frame and the crane when operating at the crossing point during the construction process. At the same time, the safe distance between the crane and the live body during the operation, and the safe distance between the capping net and the live wire and the ground wire are calculated; if it is not within the safe distance, the alarm will be immediately shut down, and the deduction will be repeated after manual adjustment.

[0132] Based on the parallel control model connecting the digital twin model of the mobile crossing frame and the actual construction of the mobile crossing frame system, parallel control of on-site construction is carried out with the construction planning scheme as the target.

[0133] S1Currently, the modeling of mobile crossing frame application scenarios is mainly decomposed from all factors such as "man-machine-object-environment". A virtual model is constructed based on the real mobile crossing frame construction scene, mainly including: geometry-physics-behavior-rule model. For the geometric dimension, according to the information such as the geometric characteristic parameters of the crossing frame, the geometric model of the equipment is constructed, including the main body of the crossing frame, the transport frame, the sealing device, etc.; for the physical dimension, according to the material properties, physical parameters and other physical characteristics of the mobile crossing frame equipment, the physical model is constructed, such as the motor power of the mobile crossing frame, the lifting range of the crossing frame, etc.; for the behavioral dimension, based on the behavioral coupling relationship between each component, a behavior and response model that describes the behavioral characteristics of the equipment is constructed, such as the crane docking with the crossing frame, and the boom lifting the crossing frame to the operating height; for the rule dimension, the rules and logic models based on the XML language are used to describe the operation and evolution of the equipment, such as adjusting the inclination angle of the crossing frame so that the crane arm can achieve the best protection area and the best impact resistance under different working conditions.

[0134] S2 mobile umbrella-shaped spanning frame knowledge graph construction description;

[0135] In the process of cross-over operation of transmission lines, the four stages of construction organization requirements, construction process specifications, construction organization structure, and resource allocation in the application scenario of mobile crossing frames are the main clues. As shown in the figure, the entity concepts and attributes concerned in each stage are analyzed to construct the knowledge graph model layer of mobile crossing frame construction. The corresponding entities and attributes are extracted from the mobile crossing frame operation construction plan, "Safety Work Regulations for Power Construction", "Standard Process for Power Transmission and Transformation Engineering of State Grid Corporation of China", "Q / GDW 11957.2-2020 Safety Work Regulations for Power Construction of State Grid Corporation of China", "DL / T5106-2017 Construction Regulations for Crossing Power Lines", etc., so as to realize the construction of knowledge graph.

[0136] The knowledge graph of the mobile spanning rack application scenario can be represented by G = (E, R, S), where E is the entity set in the knowledge base, R is the relationship set, and S is the "node-relationship-node" triple knowledge set. The data layer construction is to extract the required entities and relationships under the guidance of the pattern layer organization framework. The key technologies of data layer construction include entity extraction, knowledge fusion and knowledge update.

[0137] (1) Entity extraction: In the mobile spanning rack scene knowledge graph, each entity corresponds to a node, and the node has a label and a property:

[0138] Entity=Node:Name{Property1,Property2,·…·}

[0139] In the formula: Name represents the name of the entity Property1, Property2 represents multiple attributes of the entity, which are used to distinguish entities of different types, levels and stages. Taking the two entities of machinery and personnel as examples, knowledge extraction is performed based on the deep learning method.

[0140] The entity concepts of "machinery and equipment" and "personnel" are both noun phrases with regularity, and can be extracted through named entity recognition. This paper adopts the named entity recognition method based on the BERT model (bidirectional encoder representation of transformer) to extract entities in the text. First, the annotated text corpus constitutes a training set to train the deep learning model. The annotation mode used is the BIO mode, where B represents the beginning part of the entity, I represents the middle part of the entity, and O represents the non-entity part. Part of the text is selected, and 1,379 sentences are manually annotated after long sentence segmentation, and the training set and validation set are divided into a ratio of 9:1. The trained model can extract "machinery and equipment" and "personnel" entities in the text.

[0141] (2) Knowledge fusion: Knowledge fusion represents multiple entities with the same meaning with a single entity, that is, A1 = {a1, a2, ..., am}, where a1 ~ am represent m entities with the same meaning, and A1 is the fused entity. In the invention, there are two situations where knowledge fusion is required. ① Mixed use of full names and abbreviations. Some mechanical equipment needs to simplify their names because they are too long. For example, "mobile umbrella-shaped crossing frame" and "crossing frame" should be the same entity. ② Diversity of text representation refers to the misidentification of multiple entities due to word differences. For example, "crane" and "crane" should be the same entity. It is difficult to formulate complete knowledge fusion rules for such situations. Therefore, a text clustering algorithm is used for knowledge fusion, that is, traversing each entity, calculating the text similarity between the entity and the remaining entities, and merging entities with similarities above the threshold. The calculation formula for entity semantic similarity S is:

[0142]

[0143] Where: A1, B1 are the word frequency vectors of the two entity names after word segmentation; n is the number of word segmentations. The higher the semantic similarity S1, the higher the similarity between the two entities.

[0144] (3) Knowledge update: At present, the scenarios of mobile crossing frames are constantly changing. Therefore, the knowledge graph needs to be continuously updated to ensure the validity of the knowledge and to ensure that the knowledge graph can provide a reference for decision-making during construction planning. The form of the construction plan is relatively fixed. The main contents include project overview, crossing plan, crossing construction organization, construction technical measures, safety measures, emergency plans, civilized construction and environmental protection measures. Therefore, the model layer is relatively fixed. When adding or improving the current plan in the future, new plan knowledge will be added, so the knowledge graph data layer needs to be updated. The new construction plan must be added to the original knowledge graph after knowledge fusion. At the same time, it should be considered whether the same entity or relationship already exists in the existing graph for the newly added entity or relationship. If so, deduplication should be performed; whether there are invalid entities or relationships after the knowledge update, if so, they should be eliminated.

[0145] Digital planning and deduction description of the construction of S3 mobile umbrella-shaped spanning frame;

[0146] Construction planning: Based on the preliminary investigation and analysis of the construction site and environment to identify the main and special aspects that affect the project, combined with the knowledge graph and standardized adaptation principles constructed for the mobile spanning frame operation scenario, the relevant entity attribute information is extracted from the formed knowledge graph according to the on-site operation requirements to determine the construction methods and processes. Finally, the human resource allocation is calculated based on the operation content to generate a planning scheme.

[0147] In this patented system, the planning process can not only be given by the system automatically but also allow manual adjustment. Its main planning processes include: selection of construction methods, allocation of construction resources (machinery, personnel, tools, planned time, risk safety measures, etc.), equipment access route planning, work site layout, etc.

[0148] In the selection of construction methods, the system will give a recommended construction method in the form of conditional value comparison based on the design data of the current point, but it allows manual changes, and construction methods that do not meet the conditions will have corresponding information prompts. After selecting the construction method, enter the construction resource configuration. In this step, the system will extract the construction resource configuration such as machinery and personnel involved in the corresponding construction method from the formed knowledge graph, including the corresponding risk safety measures. The specific process is to search for the corresponding construction method entity in the knowledge graph, extract and record all the relationships of the entity after finding the entity, and arrange the results in the form of fields and display them in different system interfaces. Similarly, this step also allows on-site personnel to make manual adjustments.

[0149] See also Figure 6 In path planning, the system will automatically plan the most reasonable route based on the connected domestic road network information, and will also mark the length of the road that needs to be expanded or rebuilt, and the checkpoints that affect the passage of equipment on the route will also be automatically marked.

[0150] In the layout of the construction site, the system will set up the construction site according to the initial quadrilateral shape based on the actual surface conditions. The construction area layout can also be generated with one click, and the layout of the construction site can also be completed manually through the drag-and-drop interface operation. After the construction site is determined, the system will also calculate the corresponding compensation amount for young crops according to the quota as part of the construction budget.

[0151] (i) Project Overview

[0152] The 220kV Changdang Line #127-#137 (currently #132-#143) line reconstruction project has a total line length of 3.616km, starting from the current 220kV Changdang Line #132 and ending at the 220kV Dangjiazhuang substation. A total of 18 new towers were built, all of which were single-circuit steel tube poles, including 13 tension steel tube poles and 5 straight steel tube poles. The line conductors of this project are 2×JL / G1A-400 / 35 steel core aluminum stranded wires, and the twin conductors are arranged horizontally. The existing #132-AN1 ground wire is re-tightened and laid out; the ground wire of the new 220kV line uses two 72-core OPGW-150 optical cables throughout the line.

[0153] (ii) Construction content

[0154] The construction contents of crossing the charged 10kV Dangliu line 101HW (27#-2701 cable terminal pole); the charged 10kV Dangti line 102HW (#24~#25) are as follows:

[0155] Assemble the span frame near the AN3 tower side. After the crane is parked in place (the ground projection point 10m outside the 10kV vertical direction of the intersection of the 10kV Dangliu Line 101HW live line and 220kV is the crane rotation center), rotate the arm to a place away from the live body to install the span frame, lift the span frame and park it in place;

[0156] Assemble the spanning frame on the side of AN3-AN4 tower. After the crane is parked in place (the ground projection point 10m outside the intersection of the 10kV Dangliu Line 101HW live line and 220kV in the vertical direction of 10kV is the crane rotation center), rotate the arm to a place away from the live body to install the spanning frame, lift the spanning frame and park it in place;

[0157] Assemble the crossing frame near the side of AN4 tower. After the crane is parked in place (the ground projection point 11m outside the intersection of the 10kV party body line 102HW live line and the 220kV is the crane's rotation center), rotate the arm to a place away from the live body to install the crossing frame, lift the crossing frame and park it in place.

[0158] (iii) Crossing point parameters and sections

[0159] Taking the 12 and 28 (3#-4#) crossing live 10kV party-building lines as an example, the crossing point information is as follows:

[0160] Table 2 Crossing point parameters

[0161]

[0162] (iv) Equipment selection

[0163] According to the characteristics of this project and the span parameters, the TOPSIS method is used between the equipment performance parameters to complete the multi-objective decision-making problem, and then the corresponding knowledge is extracted in the knowledge map according to the standardized adaptation principle to complete the equipment selection and on-site resource allocation. For example, when a mobile spanning frame and crane are used for spanning operations, 8-12T cranes are generally installed with the crane. The actual number is combined with the spanning construction content to calculate the number of equipment and reasonable resource allocation.

[0164] There are two main points in selecting a mobile umbrella-shaped span: 1) Select the model based on the load of the deployed line and the parameter table of the mobile span, such as the line section and span span; 2) Select and determine the quantity based on the scope of the spanned object and the operation requirements, such as the intersection angle between the new line and the spanned object, span length, span width, and distance to the ground. Therefore, the mobile umbrella-shaped span model SKY-SY-8×12 / 50-D selected for this project can meet the operation requirements.

[0165] Table 3 SKY-SY-8×12 / 50-D spanning rack parameter table

[0166]

[0167] The selection of crane mainly considers the following three points: 1) Operation load: the deadweight of the mobile crossing frame, possible loads of accident conditions, such as the deadweight, length, span length, wire weight, and number of wire splits of the crossing frame; 2) The working range of the boom: determined by the parking point of the crane and the actual crossing point; 3) Determine the crane: According to the operation load and the working range of the boom, select the appropriate crane by looking up the table. The calculation method is as follows:

[0168] a. The weight of the mobile umbrella-shaped span is 4 tons. The impact load is 3 / 8 times the weight of the conductor of the span length according to the concentrated load borne by the network end of the network closure device in DL / T5301-2013. The operating span is 171 meters.

[0169] Impact load W JS =〔3 / 8×(171-12)+12〕×13.486 / 1000×2×1.67×1.2=3.87kN, the span frame bears a vertical load of 3.87kN;

[0170] Horizontal load P S =0.3W JS =1.161kN;

[0171] Network blocking platform load F = 0.2W JS =0.774kN;

[0172] Working wind load P W =CpA=1.7×0.625×15 2 ×11.9 / 1000=2.8kN;

[0173] Determine that the lifting capacity G of the crane at the crossing point must be greater than

[0174] b. The working range of the boom during construction near the AN3 tower is 9 meters, the length of the boom is 20 meters, and the boom angle is 59°;

[0175] The working range of the construction side arm is 11 meters, the length of the boom is 21 meters, and the boom angle is 55°.

[0176] The working range of the boom during construction near the AN4 tower is 10 meters, the length of the boom is 23 meters, and the boom angle is 59°.

[0177] The working range of the construction side arm is 12 meters, the length of the boom is 24.5 meters, and the boom angle is 55°.

[0178] c. Select two STC500E series cranes to install the spanning frame.

[0179] Appropriate human resources should be organized at the construction site of the crossing. Due to the particularity of the construction of crossing power lines, the entire construction process needs to be monitored. Each construction site is equipped with functional personnel to conduct on-site inspections, and sufficient supporting personnel are assigned to ensure safety monitoring and quality control during the construction period.

[0180] Table 4 Personnel Arrangement

[0181]

[0182] (1) Construction simulation

[0183] See also Figure 7 , 8 Based on the high-precision 3D model of the mobile spanning frame operation scene, the system automatically generates path planning and construction site layout according to the construction plan, such as the spanning frame's position in the scene, opening angle and boom height. Based on the site layout, construction process and sensor data such as wind speed sensors and height sensors, the system collects mechanical status data in real time to calculate the safe distance when operating at the spanning point for deduction.

[0184] The safety distance specifications are shown in the following table:

[0185] Table 5 Safety distances for different voltage levels

[0186]

[0187] Table 6 Minimum safe distance between the crossing frame and the object being crossed (m)

[0188]

[0189] Construction site layout: Edit and arrange various construction elements in the scene through interactive operations, such as access roads, spanning frame positions, etc. You can use different tools to adjust the position, orientation and size of construction elements. At the same time, you can customize the parameters of construction elements, such as the length of the crane boom and the lifting angle.

[0190] Construction process: Establish the predecessor and successor relationship between construction process nodes through time sequence. Adjust the basic parameters of the process node, such as the number of construction workers, construction machinery, construction cycle, etc., establish the connection between the node and the data and construction layout data in the virtual scene, and associate the process with the scene.

[0191] Simulation and deduction: The dynamic simulation of the construction plan of the mobile crossing frame of the power transmission and transformation line is realized through the deduction function. According to the setting of the construction process, the construction content of each stage in the construction process is dynamically simulated in a time sequence. During the deduction process, the spanning frame deployment angle, crane working range, arm length, angle and other parameters collected by the sensor are used to verify the safety of the spanning frame and the crane when operating at the crossing point during the construction process. At the same time, the safe distance between the crane and the live body during the operation, and the safe distance between the capping net and the live wire and the ground wire are calculated. If it is not within the safe distance, the alarm will be immediately shut down.

[0192] ① Safety verification of mobile crossing frame and crane when operating at the crossing point:

[0193] 1) Basic parameters: The mobile umbrella-shaped spanning frame of the 220kV long party line (AN3~AN4 tower) spanning the 10kV party six line 101HW (27#-2701 cable terminal pole) and the 10kV party body line 102HW (#24~#25) has a maximum amplitude of 12m for the STC500E crane, a boom length of 24.5m, and an angle of 55°. At this time, the crane boom has a lifting capacity of more than 13T, which can ensure stability when subjected to impact. The rated load bearing capacity of the SKY-SY-8×12 / 50-D spanning frame is 5T, and the frame is safe to use when subjected to impact.

[0194] 2) Stability analysis: The loads of the crane and the spanning frame when the 220kV long party line (AN3~AN4 tower) crosses the 10kV party six line 101HW (27#-2701 cable terminal pole) and the 10kV party body line 102HW (#24~#25) spanning frame are:

[0195] Deadweight of spanning frame: G1=40KN;

[0196] Accident condition impact load: main truss vertical load: WJS = 3.87kN;

[0197] Main truss horizontal load: PS = 0.3 × 8.55 = 1.161KN;

[0198] Safety net truss vertical load: F = 0.2 × 8.55 = 0.774KN;

[0199] Bring in data:

[0200] During construction, the working range of the AN3 tower crane approaching each other is 9 meters, the arm length is 20 meters, the angle is 59°, and the lifting capacity of the crane is G≥7T;

[0201] The working range of the construction side crane is 11 meters, the arm length is 21 meters, the angle is 55 degrees, and the lifting capacity of the crane is G≥6.8T;

[0202] During construction, the working range of the AN4 tower crane approaching each other is 10 meters, the arm length is 23 meters, the angle is 59°, and the lifting capacity of the crane is G≥6.9T;

[0203] The working range of the construction side crane is 12 meters, the arm length is 24.5 meters, the angle is 55 degrees, and the lifting capacity of the crane is G ≥ 6.8T;

[0204] The selected crane has a lifting capacity greater than 13T at this time and is safe to use.

[0205] ②Safety distance verification:

[0206] 1) Safety distance description: According to 7.2 of "Q / GDW 11957.2-2020 State Grid Corporation of China Electric Power Construction Safety Work Regulations Part 2: Lines", the safety distance between cranes and hanging parts and 10kV live parts is 1.5m in the horizontal direction, and the safety distance between cranes and hanging parts and 110kV live parts is 4m in the horizontal direction. According to 3.2.7 of "DL / T5106-2017 Construction Regulations for Crossing Power Lines", the safety distance between the capping net and the 10kV line is 1.5m, and the safety distance between the capping net and the 110kV ground wire is 1m.

[0207] Therefore, the horizontal safety distance of the crane arm for 10kV live parts is required to be no less than 4.5m, the horizontal safety distance of 110kV live parts is required to be no less than 6.5m, and the vertical distance between the umbrella-shaped spanning frame and the ground wire / optical cable is required to be no less than 2m.

[0208] 2) Calculation of safety distance: During the process of positioning the crossing frame, the live line is projected onto the ground, and a pennant soft fence is inserted at the projection. Tie a Φ6 Dyneema rope at 10m of the first boom of the crane, and tie a weight at the lower end of the rope. According to the crane boom angle α, the vertical and horizontal safety distance between the boom and the 10kV line is 4.5m, the vertical and horizontal safety distance between the boom and the 110kV line is 6.5m, the vertical and horizontal distance L1 between the rotation center of the crane parking place and the live wire, and the height H between the conductor on the live wire and the ground, ensure that the vertical distance L between the weight and the pennant soft fence is ≥, to prevent the crane from accidentally entering the live area during the rotation to the working position. When crossing the live line, the angle β is, in principle, perpendicular to the line when the boom rotates to the working position. At this time, the working range of the boom is the smallest when ensuring the safety distance, and when the crane is placed in the designed position, the boom will not accidentally enter the live area during the entire rotation process.

[0209] In summary, taking the construction scenario of a mobile crossing frame for a power transmission and transformation line as an example, the digital twin scenario is used as the basis for planning in the artificial system. The data collected by the sensor is used to perform real-time measurement and deduction of the safety distance during the operation, and the deduction is used as a parallel system calculation method to provide construction personnel with an analysis of the optimal control solution and support accurate decision-making. At the same time, the results of the planning and deduction are used as the optimal control target state to be achieved by parallel control.

[0210] Through modern control theory, the real motion control of the equipment is completed, that is, the parallel control of the equipment is realized. Taking the motion control of the crossing frame equipment as an example, parallel control calculation is carried out. The overall operation process of the crossing frame can be abstracted and summarized into the following actions. First, the crane reaches the specified position, then the legs are extended, the boom is lifted and rotated to the specified angle, and the umbrella of the crossing frame is fully opened. According to the spatial position of the wire, the boom rotates to the working position and then adjusts the pitch angle of the crossing frame to make it horizontal to complete the task of opening the crossing frame. When the wire operation is completed, the crossing frame is folded in the opposite direction according to the same action. For the specific action of boom lifting, when calculating according to modern control, the first step is to construct the state equation for the boom. The constant matrix in the equation represents the various size information of the boom, that is, the action transfer relationship between different motion variables. The second step is to propose the cost function of the parallel control of the error adjustment problem, and rely on the Bryson method to calculate the state cost weighting matrix Q and the control cost weighting matrix R. Finally, the linear quadratic controller is used to solve the state equation on the basis of minimizing the cost function, and the optimal response of the control signal to the target is obtained.

[0211] Optimal control tasks can be divided into two categories according to their objectives: state regulation problems and trajectory tracking problems. The main difference lies in the different objective functions.

[0212] The state regulation problem refers to the problem of making the system state converge to a desired state by controlling the input under a given initial state. Usually, the objective function of the state regulation problem is to minimize the error between the system state and the desired state. This error is the cost function J, and by minimizing J, the system state converges to the desired state.

[0213] The tracking problem is to make the system state follow the desired trajectory through control input under a given desired trajectory. Usually, the objective function of the tracking problem is to minimize the error between the system state and the desired trajectory. The goal of the tracking problem is to minimize the cost function so that the system state follows the desired trajectory.

[0214] Therefore, the main difference between the state regulation problem and the tracking problem lies in the difference in the desired state. In the state regulation problem, the desired state is a fixed state, and the goal of the control input is to make the system state converge to the desired state. In the tracking problem, the desired trajectory is a time function, and the goal of the control input is to make the system state follow the desired trajectory.

[0215] Taking the control of the parachute arm of the straddle rack as an example, this problem belongs to the tracking problem, that is, it is hoped that the error between the straddle rack state at any time and the expected trajectory is minimized. However, the computing resources consumed by the tracking problem in the calculation process are quite large. Therefore, the time of the tracking problem can be discretized, and for a certain moment, the problem is transformed into a state adjustment problem.

[0216] In the classical control equation, the control method mainly consists of the state equation and the control equation:

[0217] Equation of state:

[0218] Control equation: u = h(x);

[0219] The control method of the parallel control equation is as follows:

[0220] Equation of state:

[0221] Governing equations:

[0222] Among them, x is the state vector, u is the control vector, f and h represent the state function and control function;

[0223] The unified equation is expressed as:

[0224] Where z = [x T ,u T ] T ,G(·)=[f(·) T , h(·) T ]; the equation implicitly contains an ordinary differential equation system with independent variable time t, which is a time-independent self-consistent system, which provides a new idea for the nonlinear parallel control problem; on the other hand, the form of the self-consistent system establishes a connection between parallel control and neural networks based on kernel functions, and the Shenjiang network can be used to solve nonlinear parallel control equations;

[0225] Considering the advantages of parallel systems in dealing with highly complex systems, applying classical control equations to physical space and realizing control of the real world according to classical control principles, the implementation of parallel control equations in information space means that complex problems can be calculated in the cloud. In summary, it is "simple physical space, complex information calculation". This also reflects the theoretical characteristics of parallel systems of "expanding and governing"; the control vector u of the parallel control equation can be used as the setting target of the classical control equation, linking the physical system with the artificial system to achieve consistency in goals.

[0226] By discretizing the time of the tracking problem, for a certain moment, the problem is transformed into an error adjustment problem, and the error adjustment problem is transformed into a type of system adjustment problem:

[0227] Control objectives are divided into two categories. One is the regulation problem, which is to ensure the stable operation of the target. The target is usually The position and velocity are 0. Another type is the tracking problem, which is to ensure that the controlled target can move along the trajectory we set. or etc. Since the spanning frame is a full-process control rather than a target control, the following control is carried out for the trajectory tracking problem of the spanning frame. Considering the high degree of freedom of the actual spanning frame, limited by time and computing costs, the following case study is only carried out with an idealized two-dimensional umbrella arm. In order to reflect the parallel control idea, this study adopts the linear quadratic controller (Linaer Quadratic Regulator, LQR), which is a typical modern control method, similar to adaptive dynamic programming (Adaptive Dynamic Programming, ADP), differential dynamic programming (Differential Dynamics Programming, DDP) and other methods, and can use twin information for real-time parallel control.

[0228] The parallel control problem of the straddle rack is essentially the target tracking problem of the target speed and position of the straddle rack. The time of the target tracking problem is discretized. For a certain moment, the target expected state of the straddle rack at time t+1 is:

[0229] x d [t+1]=A d x d [t]

[0230] Among them A d is the target transfer matrix, x d [t] is the expected state of the spanning rack at time t, x d [t+1] is the expected state of the spanning rack at time t+1; construct the spanning rack state equation: x[t+1]=Ax[t]+Bu[t]

[0231] Among them, x[t+1] is the state target of the spanning rack at time t+1, A is the state matrix of the spanning rack, B is the input control matrix of the spanning rack, and u[t] is the input control vector of the spanning rack system at time t;

[0232] Since the passive target velocity and posture of the straddle frame are both constant vectors, A d =I; Definition error:

[0233] e[t]=x[t]-xd [t]

[0234] Where e[t] is the control error of the spanning frame at time t;

[0235] At this point, the target tracking problem is transformed into a type of error adjustment problem; integrating the above formula:

[0236]

[0237] At this time, the augmented form space state equation is obtained: a [t+1]=A a x a [t]+B a u[t]

[0238] where x a [t+1] is the state vector of the spanning rack at time t+1 after merging, A a is the state matrix across the frame after merging, x a [t] is the state vector of the spanning rack at time t after merging, B a is the input control matrix of the merged spanning rack; then:

[0239]

[0240] Among them, C a [II]; relying on parallel control theory, the cost function J can be constructed:

[0241]

[0242] Among them, S is a symmetric positive definite matrix, which is usually used to weigh the importance of the state error e[t] at time t, Q is the state cost weighting matrix, R is the control cost weighting matrix, and N represents the previous N moments;

[0243] This cost function is the cost function construction in the tracking problem: represents the error between the current state and the expected state at the current time; e[t] T Qe[t] represents the cost or energy required to reduce the error accumulated in the past; u[t] T Ru[t] is the energy required for the system to reach the next state at this moment;

[0244] Substituting e[t] into J, we get:

[0245]

[0246] Let S a =C a T SC a , Qa =C a T QC a , the cost function is:

[0247]

[0248] At this time, the cost function and the augmented form space state equation conform to the linear quadratic standard form; through the above process, a type of error tracking problem is successfully converted into a type of regulation problem;

[0249] The goal of the linear quadratic controller is to design a controller that provides a state feedback u[t] = -Kx a [t] to minimize the cost function, we get:

[0250]

[0251] Assume a constant matrix P that satisfies:

[0252]

[0253] Take K = R -1 B a T P, then A a T P+PA a +Q a -PB a R -- B a T P=0This equation is the Riccati equation;

[0254] The calculation process of K: First, the state cost weighted matrix Q is determined by the Bryson method a and the control cost weight matrix R, Q a is a semi-positive definite matrix, and R is a positive definite matrix; then, the Riccati equation is solved to obtain the matrix P;

[0255] The goal of LQR is to select the control input u[k] so that the cost function J is minimized. The weight matrices Q and R are used to adjust the importance of different states and control inputs to achieve better system control. In the specific control process, we only need to determine the matrix S a , Q a and R; the following is the numerical calculation process:

[0256] Define the state equation of the parachute arm:

[0257]

[0258] Among them, α, q, and θ represent the acceleration, velocity, and angle of the umbrella arm during motion, respectively. The input δ is the motor control deflection angle, and the output variable y is the umbrella arm pitch angle. Here, δ is our u[k].

[0259] It is necessary to determine the state cost weighting matrix Q and the control cost weighting matrix R. Generally, the selection of these two types of matrices depends on the Bryson method. Here we choose: R = 1; this is because The calculation process is:

[0260]

[0261]

[0262] Here, Q i,i is the element in the i-th row and i-th column of the Q matrix, the maximum acceptable value of (error states ) 2 is the square of the maximum acceptable value of the state error, and n is the number of states.

[0263]

[0264]

[0265] Here, R j,j is the element in the jth row and jth column of the R matrix, the maximum acceptable value of (error inputs ) 2 is the square of the maximum acceptable value of the control input, and m is the number of states. Since the target is a single target, R is a 1-dimensional matrix, and since the input is three variables, Q is a 3-dimensional matrix: [u, S, P] = lqr (A, B, Q, R);

[0266] Calculate the optimal gain u[k], the solution S of the associated algebraic Riccati equation and the poles P of the closed-loop system are:

[0267]

[0268]

[0269]

[0270] Finally, calculate K = R -1 B a T P.

[0271] Embodiment 2:

[0272] See also Figure 2A mobile umbrella-type spanning frame parallel control system, the system is used to execute the aforementioned mobile umbrella-type spanning frame parallel control method, specifically including:

[0273] Digital twin and parallel control model construction module: used to build a digital twin model of the construction scene based on the real construction scene, build digital twin models of each mobile crossing frame based on different crossing frame models, and build a parallel control model connecting the digital twin model of the mobile crossing frame and the mobile crossing frame system under actual construction;

[0274] Knowledge graph construction module: used to construct a knowledge graph of mobile umbrella-shaped spanning frames and construction operation scenarios based on the design and construction plan of the line to be constructed, the actual construction plan of the historical construction line, and the current construction safety specifications and standards;

[0275] Construction planning and deduction model building module: used to build the construction planning of the mobile umbrella-shaped spanning frame based on the digital twin model and knowledge graph, and to perform construction deduction in the digital twin model to obtain the construction planning scheme;

[0276] Parallel control implementation module: It is used to perform parallel control of on-site construction based on the parallel control model connecting the digital twin model of the mobile crossing frame and the actual construction of the mobile crossing frame system, with the construction planning scheme as the target.

[0277] Embodiment 3:

[0278] See also Figure 3 A mobile umbrella-type spanning frame parallel control device comprises a memory and a processor, wherein the memory is used to store computer program codes and transmit the computer program codes to the processor; the processor is used to execute the aforementioned mobile umbrella-type spanning frame parallel control method according to the instructions in the computer program codes.

[0279] Embodiment 4:

[0280] A computer-readable storage medium stores a computer program, which is executed by a processor to implement the aforementioned parallel control method for a movable umbrella-shaped spanning frame.

Claims

1. A parallel control method for a mobile umbrella-shaped spanning frame, characterized in that: The steps include: Build a digital twin model of the construction scene based on the real construction scene, build digital twin models of each mobile crossing frame based on different crossing frame models, and build a parallel control model connecting the digital twin model of the mobile crossing frame and the mobile crossing frame system under actual construction; Based on the design and construction plan of the line to be constructed, the actual construction plan of the historical construction line, and the current construction safety specifications and standards, a knowledge graph of mobile umbrella-shaped spanning frames and construction operation scenarios is constructed; Based on the digital twin model and knowledge graph, a construction plan for a mobile umbrella-shaped spanning frame is constructed, and construction simulation is performed in the digital twin model to obtain a construction plan. Based on the parallel control model connecting the digital twin model of the mobile crossing frame and the actual construction of the mobile crossing frame system, parallel control of on-site construction is carried out with the construction planning scheme as the target.

2. A method for parallel control of a mobile umbrella-shaped spanning frame according to claim 1, characterized in that: A digital twin model of the construction scene is constructed based on the real construction scene, and the relevant parameter information of the line to be constructed is imported into the model, including the distribution coordinate information of each tower, tower model, tower number and layout and installation design information of the transmission line; at the same time, the domestic road network information based on the map is synchronized in proportion on the bottom layer of the model, so that the coordinates of the digital twin scene correspond to the coordinates of the map one by one, and a digital twin model of the construction scene is obtained; Based on different types of crossing frames, digital twin models of mobile crossing frames are constructed. A three-dimensional geometric model is obtained through parametric design. The physical, behavioral and rule models of each subsystem are defined according to the actual operation mode of the mobile crossing frame to form a high-fidelity mapping of the physical entity. Then, digital twin models are built for the construction equipment and materials that are used in conjunction with the spanning frame. The constructed digital twin model of the mobile spanning frame and the digital twin models of the construction equipment and materials can all be arranged in the digital twin model of the construction scene according to the construction plan; A parallel control model connecting the digital twin model of the mobile crossing frame and the mobile crossing frame system under actual construction is constructed, so that the control signal of the digital twin model of the mobile crossing frame can synchronously control the mobile crossing frame system under actual construction. At the same time, the mobile crossing frame system under actual construction will feed back the received mobile crossing frame sensor signal to the digital twin model of the mobile crossing frame to assist the feedback adjustment of the parallel control model.

3. A parallel control method for a mobile umbrella-shaped spanning frame according to claim 1, characterized in that: The method for constructing a parallel control model connecting the digital twin model of the mobile crossing frame and the mobile crossing frame system in actual construction is as follows: The parallel control problem of the straddle rack is essentially the target tracking problem of the target speed and posture of the straddle rack. The time of the target tracking problem is discretized. For a certain moment, the target expected state of the straddle rack at time t+1 is: x d [t+1]=A d x d [t] Among them A d is the target transfer matrix, x d [t] is the expected state of the spanning rack at time t, x d [t+1] is the expected state of the spanning rack at time t+1; Construct the spanning frame state equation: x[t+1]=Ax[t]+Bu[t] Among them, x[t+1] is the state target of the spanning rack at time t+1, A is the state matrix of the spanning rack, B is the input control matrix of the spanning rack, and u[t] is the input control vector of the spanning rack system at time t; Since the passive target velocity and posture of the straddle frame are both constant vectors, A d =I; Definition error: e[t]=x[t]-x d [t] Where e[t] is the control error of the spanning frame at time t; At this point, the target tracking problem is transformed into a type of error adjustment problem; integrating the above formula: At this time, the augmented form space state equation is obtained: x a [t+1]=A a x a [t]+B a u[t] where x a [t+1] is the state vector of the spanning rack at time t+1 after merging, A a is the state matrix across the frame after merging, x a [t] is the state vector of the spanning rack at time t after merging, B a is the input control matrix of the merged spanning rack; then: Among them, C a for [II]; Relying on parallel control theory, the cost function J can be constructed: Among them, S is a symmetric positive definite matrix, which is usually used to weigh the importance of the state error e[t] at time t, Q is the state cost weighting matrix, R is the control cost weighting matrix, and N represents the previous N moments; This cost function is the cost function construction in the tracking problem: represents the error between the current state and the expected state at the current time; e[t] T Qe[t] represents the cost or energy required to reduce the error accumulated in the past; u[t] T Ru[t] is the energy required for the system to reach the next state at this moment; Substituting e[t] into J, we get: Let S a =C a T SC a , Q a =C a T QC a , the cost function is: At this time, the cost function and the augmented form space state equation conform to the linear quadratic standard form; through the above process, a type of error tracking problem is successfully converted into a type of regulation problem; The linear quadratic controller uses state feedback u[t] = -Kx a [t] to minimize the cost function, we get: Assume a constant matrix P that satisfies: Take K = R -1 B a T P, then A a T P+PA a +Q a -PB a R -1 B a T P=0This equation is the Riccati equation; The calculation process of K: First, the state cost weighted matrix Q is determined by the Bryson method a and the control cost weight matrix R, Q a is a semi-positive definite matrix, and R is a positive definite matrix; then, solve the Riccati equation to obtain the matrix P; finally, calculate K = R -1 B a T P.

4. A method for parallel control of a mobile umbrella-shaped spanning frame according to claim 1, characterized in that: Based on the design and construction plan of the line to be constructed, the actual construction plan of the historical construction line, and the current construction safety specifications and standards, a knowledge graph of mobile umbrella-shaped spanning frames and construction operation scenarios is constructed; In the process of knowledge graph construction, construction organization requirements, construction process specifications, construction organization structure, and resource allocation are the main clues; Knowledge update: Based on the newly generated or modified construction plan of the mobile umbrella-shaped spanning frame construction planning and simulation model, the knowledge graph is updated and stored. The newly added construction plan needs to be added to the original knowledge graph after knowledge fusion. At the same time, it should be considered whether the same entity or relationship already exists in the existing knowledge graph for the newly added entity or relationship. If so, deduplication should be performed; whether there are invalid entities or relationships after the knowledge update, if so, they should be eliminated; For modifications to construction safety specifications and standards, the knowledge graph is updated and stored. For data where relevant provisions in new construction safety specifications and standards conflict with past provisions, the knowledge graph generates a comparison list and determines it manually.

5. The parallel control method of a mobile umbrella-shaped spanning frame according to claim 1 is characterized in that: Based on the digital twin model and knowledge graph, a construction plan for a mobile umbrella-shaped spanning frame is constructed, and construction deduction is performed in the digital twin model to obtain a construction plan: the relevant information parameters in the digital twin model of the construction scene are input into the knowledge graph. The knowledge graph generates an initial construction plan based on the construction points of the construction plan for the line to be constructed. In the generation process, the actual construction plans of similar construction points in the historical construction line and the current construction safety specifications and standards need to be considered; The initial construction plan generated by the knowledge graph is adjusted based on the actual situation. The knowledge graph corrects the number or parameters of related entities based on the adjustment of the construction plan to improve the construction plan. The improved construction plan is input into the digital twin model. The digital twin model calls the spanning frame, construction equipment, and construction materials in the construction plan and arranges them in the digital twin model of the construction scene. The spanning frame, construction equipment, and construction materials are manually arranged and adjusted, and then the construction simulation of the spanning frame is started. If the simulation is successful, the construction plan is obtained. If the simulation fails, the cause of the failure is analyzed, and the parameters are adjusted before re-simulation.

6. A method for parallel control of a mobile umbrella-shaped spanning frame according to claim 5, characterized in that: During construction simulation: Carry out safety verification calculations for the post-construction planning to confirm that the construction load-bearing capacity meets the construction standard requirements; Based on the high-precision three-dimensional model of the mobile crossing frame operation scene, the deduction function is used to realize the dynamic simulation of the construction plan of the mobile crossing frame of the transmission and transformation line. According to the setting of the construction process, the construction content of each stage in the construction process is dynamically simulated in a time sequence manner; in the deduction process, the spanning frame deployment angle, crane working range, arm length, angle and other parameters collected by the sensor are used to verify the safety of the spanning frame and the crane when operating at the crossing point during the construction process. At the same time, the safe distance between the crane and the live body during the operation, and the safe distance between the capping net and the live wire and the ground wire are calculated; if it is not within the safe distance, the alarm will be immediately shut down, and the deduction will be repeated after manual adjustment.

7. A method for parallel control of a mobile umbrella-shaped spanning frame according to claim 5 or 6, characterized in that: The construction planning and construction simulation are both carried out in the digital twin model. The digital twin model presents the digital twin scene with planned construction in the form of a background display, and the generation results and modifications of the construction planning are displayed in the form of a catalog list. The construction simulation is displayed in the digital twin scene in the form of a proportional restoration of the digital twin spanning frame, the construction equipment and materials supporting the spanning frame.

8. A mobile umbrella-shaped spanning frame parallel control system, characterized in that: The system is used to execute the parallel control method of the mobile umbrella-shaped crossing frame according to any one of claims 1 to 7, specifically comprising: a digital twin, a parallel control model construction module, a knowledge graph construction module, a construction planning and deduction model construction module, and a parallel control implementation module; the digital twin, parallel control model construction module: used to construct a construction scene digital twin model based on a real construction scene, to construct digital twin models of each mobile crossing frame based on different crossing frame models, and to construct a parallel control model connecting the digital twin model of the mobile crossing frame and the mobile crossing frame system under actual construction; Knowledge graph construction module: used to construct a knowledge graph of mobile umbrella-shaped spanning frames and construction operation scenarios based on the design and construction plan of the line to be constructed, the actual construction plan of the historical construction line, and the current construction safety specifications and standards; Construction planning and deduction model building module: used to build the construction planning of the mobile umbrella-shaped spanning frame based on the digital twin model and knowledge graph, and to perform construction deduction in the digital twin model to obtain the construction planning scheme; Parallel control implementation module: It is used to perform parallel control of on-site construction based on the parallel control model connecting the digital twin model of the mobile crossing frame and the actual construction of the mobile crossing frame system, with the construction planning scheme as the target.

9. A mobile umbrella-shaped spanning frame parallel control device, characterized in that: The method comprises a memory and a processor, wherein the memory is used to store computer program code and transmit the computer program code to the processor; The processor is used to execute the parallel control method of the mobile umbrella-shaped spanning frame according to any one of claims 1 to 7 according to the instructions in the computer program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which is executed by a processor to implement the method for parallel control of a mobile umbrella-shaped spanning frame as described in any one of claims 1 to 7.

Images