Search

RU-2861693-C1 - METHOD FOR DETERMINING ROUTES FOR MOVING INDUSTRIAL PLANTS AND SYSTEM FOR ITS IMPLEMENTATION

RU2861693C1RU 2861693 C1RU2861693 C1RU 2861693C1RU-2861693-C1

Abstract

FIELD: navigation. SUBSTANCE: invention relates to the field of logistics resource and/or vehicle distribution, route planning, and discloses a system and a method for moving industrial installations and services that support these installations to the corresponding locations. The method for determining routes for moving industrial installations includes the following steps: receiving a request to determine routes for moving industrial installations, which contains the coordinates of the locations, the type of industrial installations and the schedule of visits to locations, the presentation of the initial data in tensor and/or matrix form, the construction of a matrix of target indicators, a tensor of permissible movements, and a tensor of movement times, the formulation of the problem based on the initial data and the resulting matrix and tensors, and the submission of the problem to a computing device for finding its optimal solution to determine the routes for moving the industrial installations. EFFECT: technical result is to increase the speed of route planning for tasks with a large number of locations and agents, taking into account restrictions on the type of moving agent and/or locations. 10 cl, 7 dwg

Inventors

  • Lezhnev Konstantin Eduardovich
  • Ismagilov Niyaz Salavatovich
  • Varzegov Evgenij Vladimirovich
  • KUZNETSOV KIRILL ALEKSANDROVICH
  • Mamaev Ilya Sergeevich
  • Maksyutin Aleksandr Valerevich
  • Ivanov Yurij Valerievich
  • Durova Anastasiya Vyacheslavovna

Dates

Publication Date
20260507
Application Date
20250528

Claims (20)

  1. 1. A system for determining the routes of movement of industrial installations, containing
  2. - at least one processing tool configured to process initial information containing data including at least the coordinates of locations, the type of industrial installations and a schedule of visiting locations, representing the initial data in tensor and/or matrix form, constructing a matrix of target indicators, a tensor of permissible movements and a tensor of movement times;
  3. - a data storage module configured to store at least the initial information;
  4. - at least one route determination tool, configured to formulate at least one task taking into account the initial data and the obtained matrices and tensors and to direct said at least one task to the computing device;
  5. - a computing device designed with the ability to receive said at least one task, search for its optimal solution for determining the routes for moving industrial installations, wherein the optimal solution obtained corresponds to the minimization of the target indicator selected by the planning operator;
  6. - at least one post-processing tool configured to convert the obtained optimal solution into a user-interpretable format, graphically display the said optimal solution, and transmit the data of the determined routes to the user.
  7. 2. The system according to claim 1, which comprises at least one correlation tool configured to receive dynamic location data from user devices, correlate said data received from user devices with data from the storage module, and provide information about the upcoming and current route.
  8. 3. The system according to item 1, which is configured to communicate with the central system or the system user interface, and/or the scheduling operator interface via a private virtual, and/or local, and/or public network, and
  9. in which the computing device is a remote and/or cloud server, and/or an analog computing accelerator, and/or a process in a virtual machine on a server or user device.
  10. 4. A method for determining the routes for moving industrial installations, which includes at least the following steps:
  11. receiving a request in which a request is received by the route determination system to determine the routes for the movement of industrial installations, containing initial information containing data that includes at least the coordinates of the locations, the type of industrial installations and the schedule of visits to the locations;
  12. an initial stage in which the initial information is processed using at least one processing tool and a representation of the initial data is formed in tensor and/or matrix form, a matrix of target indicators, a tensor of permissible movements and a tensor of movement times is constructed, and the processed initial data is stored in a storage module;
  13. a calculation stage in which, using at least one route determination tool, at least one task is formulated taking into account the initial data and the obtained matrices and tensors, after which said at least one task is sent to a computing device, by means of which a search is performed for its optimal solution for determining the routes for the movement of industrial installations, wherein the optimal solution obtained corresponds to the minimization of the target indicator selected by the planning operator;
  14. the post-processing stage of the solution, which involves converting the obtained optimal solution into a user-interpretable format and graphically displaying the specified optimal solution, and ensuring the transfer of data on specific routes to the user.
  15. 5. The method according to claim 4, in which, at the initial stage, after constructing the matrix of target indicators, the tensor of admissible movements and the tensor of movement times, clustering of data is performed to reduce the size of said at least one problem, wherein the clustering can be performed by redistributing the preliminary solution obtained on the basis of a greedy algorithm and using standard algorithms with the condition of satisfying constraints within each of the clusters.
  16. 6. The method according to claim 4, wherein at least one said problem is an integer linear programming problem, the optimal solution of which is found using the branch and bound method.
  17. 7. The method according to paragraph 4, in which, to solve said at least one problem, quadratic binary optimization without constraints is used based on the formulated problem of integer quadratic programming, wherein said at least one problem is sent to a computing device, which is an analog device, and a search is performed for an optimal solution to said at least one problem for determining the routes of movement of industrial installations by means of the analog device.
  18. 8. The method of claim 7, wherein the computing device is an analog computing accelerator that is one of the following: a coherent Ising machine, an FPGA system, an ASIC, a quantum computer, or another analog accelerator.
  19. 9. The method according to paragraph 4, in which at the initial stage a request is sent to a geographic information system to obtain information about roads on the ground, in the case of incompleteness of which graph linking methods are used, and in the case of complete absence of which roads are constructed between locations along a straight line on a sphere using at least the formula
  20. where L is the distance between two locations on the Earth's surface; R is the average radius of the Earth, equal to 6371.3 km, θ 1 , θ 2 are the latitude of 1 and 2 locations, respectively, φ 1 , φ 2 are the longitude of 1 and 2 locations, respectively.

Description

AREA OF TECHNOLOGY This invention relates to the field of logistical resource and/or vehicle distribution and route planning, and discloses a system and method for relocating industrial installations and the services supporting these installations to appropriate locations. The presented method and system are not limited to industrial installations and sites; they can be used to plot routes over large areas with a specific work schedule, for example, during land development and/or work on large construction sites, as well as in agricultural areas. LEVEL OF TECHNOLOGY Key characteristics required for determining industrial plant routes include the ability to plan a complex route for a large number of locations, for example, at least 50 locations, and a certain number of production plants, for example, at least 15 plants, in accordance with target parameters such as route length, route time, accident probability, etc.; the ability to work with fixed time windows for each location, meaning there are strict start and end times for servicing a given location; the ability to calculate the best route from one location to any other using regional road data or, if such data is unavailable, an approximate route using other estimates; the need to take into account differences in the technical design of production plants and their limitations in servicing certain locations. Failure to meet these conditions may result in non-productive equipment downtime. Due to the large number of locations planned for servicing, optimal assignment of production units can be significantly complex and cannot be solved using standard analytical methods in a reasonable time—that is, time sufficient to recalculate several different assignment options during the planning period. For example, if a week is allocated for assignment planning, a reasonable time could be considered up to 1 hour. If one day is allocated for planning, then up to 1 minute. Traditionally, methods for solving such problems consist of several stages, which include collecting the necessary data on the coordinates of the locations to be visited, the service schedule, and the road graph in the spatial region of interest, using the method to solve the problem of planning agent routes in accordance with the selected target landmarks, and informing agents about the constructed routes. In this case, agents are understood to be any objects whose routes need to be planned within the framework of the method. That is, agents can include, but are not limited to: a person plotting a route in a city or off-road conditions; Machines that must visit multiple locations, for example, to deliver within a specified timeframe; various types of special equipment, in particular, cranes, bulldozers, dump trucks, combine harvesters, and tractors; industrial installations, including drilling rigs, that must visit specific locations, such as drilling pads, well pads, construction sites, and other industrial locations, etc. Technical limitations or technical characteristics that must be taken into account during calculations may include, for example, the rig's lifting capacity, the presence of a top drive, and the type of drilling rig (mobile, echelon, or stationary). Other technical characteristics of the drilling agents may also serve as technical limitations, depending on the type of drilling agent and the technical field in which the route is determined. Below, methods for route planning are described in detail and an overview of current methods for solving this problem is offered, as well as analogs of the claimed system and method are considered. Planning methods can be divided into heuristic and exact methods. Heuristic methods do not guarantee finding the best solution, such as the shortest route, but they are usually significantly faster. An example of a heuristic method is the greedy method, which "sends" the agent to the nearest unvisited location at each subsequent step. Another example is the simulated annealing method, which is based on simulating the physical annealing process that occurs during the crystallization of matter. Exact methods allow one to accurately solve the problem and find the best solution, such as the shortest route. They typically take significantly longer than heuristic methods, but are guaranteed to yield better results. An example of an exact method is an exhaustive search of all possible routes and selecting the shortest one. Heuristic methods are also divided into deterministic and stochastic. Deterministic methods imply a strict set of operations and, upon completion, produce an answer. Deterministic methods will always produce the same answer to a given problem. However, their "heuristic" nature implies that this answer will not be the best. Stochastic methods operate in several steps, with a randomly selected route change at each step. Over the course of several runs, stochastic methods can produce different solutions to the same problem. Moreover, the so