Intelligent power plant joint inspection system and method based on intelligent robot

Abstract

The invention discloses an intelligent power plant joint inspection system and method based on an intelligent robot, and relates to the technical field of circuit inspection. Path information in an intelligent power plant area is collected in real time, an inspection network is constructed based on the collected path information, and meanwhile, a coordinate system and a reference point are set based on the constructed inspection network; based on the set coordinate system and the reference point, the intelligent robot in the operation process is accurately positioned in real time; and after the positioning is completed, establishing a real-time prediction digital twinborn model through a data simulation algorithm process based on accurate positioning of the intelligent robot, setting of a coordinate system and reference points, performing path planning on the constructed inspection network through a path planning algorithm, and inputting a path planning result into the real-time prediction digital twinborn model. And finally, the intelligent robot is controlled based on the established real-time prediction digital twinborn model, the path planning result and the current accurate positioning of the intelligent robot, so that the inspection efficiency of the power plant is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of circuit inspection, and in particular to a smart power plant joint inspection system and method based on intelligent robots. Background Art

[0002] Traditional inspection methods are usually manual inspections, which require inspectors to have high professional capabilities to deal with various situations. However, due to the manual nature of traditional inspection methods, the inspection efficiency is low and the inspection accuracy is insufficient.

[0003] The prior art, such as the invention patent application with announcement number: CN116455083A, discloses a patrol control system for smart grids, the method of which includes: through the generation of patrol plans, reception of patrol plans by patrol experts, on-site patrols and uploading of patrol results, this orderly online management mechanism realizes patrol process control and improves the smooth progress of safety patrol work. By setting up offline patrol shields as a medium for receiving patrol plans and storing and uploading patrol results, on the one hand, it makes it difficult for other external personnel to rashly receive patrol plans and change patrol result data. On the other hand, patrol experts need to carry patrol shields for on-site patrols and upload patrol locations and patrol results in sequence. If the patrol results are not recorded, the next patrol location cannot be uploaded, thereby improving the quality of power grid patrol work.

[0004] As can be seen from the above scheme, the current inspection method is still through traditional manual inspection. The efficiency of power grid inspection is low, and the inspection lacks real-time performance and accuracy, which has certain limitations. Summary of the Invention

[0005] The purpose of the present invention is to provide a smart power plant joint inspection system and method based on intelligent robots, which solves the problems existing in the background technology.

[0006] To solve the above technical problems, the present invention adopts the following technical solution: The present invention provides a smart power plant joint inspection method based on intelligent robots, which specifically includes the following steps:

[0007] S1. Collect path information within the smart power plant area in real time and build an inspection network based on the collected path information;

[0008] S2. Set the coordinate system and reference points based on the established inspection network, and accurately position the intelligent robot in real time based on the set coordinate system and reference points;

[0009] S21. Setting a coordinate system and reference points based on the established inspection network;

[0010] S22. Establish an intelligent robot motion model and an intelligent robot reference model based on the constructed inspection network;

[0011] S23, based on the established intelligent robot motion model and intelligent robot reference model, through an iterative positioning calculation method, completing the precise positioning of the intelligent robot during operation;

[0012] S3. Based on the precise positioning of the intelligent robot, setting of coordinate systems and reference points, a real-time prediction digital twin model is established through a data simulation algorithm process;

[0013] S4. Path planning is performed on the constructed inspection network using a path planning algorithm, and the path planning results are input into the real-time prediction digital twin model;

[0014] S5. Control the intelligent robot based on the established real-time prediction digital twin model, path planning results and the current precise positioning of the intelligent robot.

[0015] Preferably, the real-time collection of path information within the smart power plant area and the construction of an inspection network based on the collected path information include the following steps:

[0016] Set up smart power plant areas and collect path information within the smart power plant areas in real time;

[0017] Set path information m By a triple {a m ,z m ,d m}composition, a m Indicates the starting point of the path information, z m Indicates the end point of the path information, d m Indicates the distance of the path;

[0018] The collected path information is aggregated to build an inspection network.

[0019] Preferably, the setting of the coordinate system and reference points based on the constructed inspection network comprises the following steps:

[0020] The center area of ​​the smart power plant is set as the coordinate origin, and the horizontal direction of the smart power plant area is used as the X-axis of the three-dimensional coordinate system, the longitudinal direction is used as the Y-axis of the three-dimensional coordinate system, and the vertical direction is used as the Z-axis of the three-dimensional coordinate system; at the same time, based on the constructed three-dimensional coordinate system, m, n, and b reference points are evenly set in the directions of the three axes according to unit distances.

[0021] Preferably, the establishing of the intelligent robot motion model and the intelligent robot reference model based on the constructed inspection network comprises the following steps:

[0022] The motion model of the intelligent robot is as follows:

[0023] x t =f(x t-1,c t-1,i )+θ t ;

[0024] Among them, x t Indicates the position of the intelligent robot in the set coordinate system at time t, x t-1 represents the position of the intelligent robot in the set coordinate system at time t-1, c t-1,i Indicates that the intelligent robot monitoring device is at x at time t-1 t-1 Position relative to reference point y i The measurement data of f(x t-1 ,c t-1,i ) represents the position x t-1 and measurement data c t-1,i The motion equation of the intelligent robot, θ t represents the motion error of the intelligent robot at time t, which is the error distance between the set position and the actual position;

[0025] The reference model of the intelligent robot is as follows:

[0026] c t,i =g(x t ,y i )+θ t,i ;

[0027] Among them, c t,i Indicates that the intelligent robot monitoring device is at x at time t t Position relative to reference point y i The measurement data, y i Indicates the i-th reference point in the set coordinate system, g(x t ,y i ) represents the position x t and reference point y i The reference equation of the intelligent robot, θ t,i Represents the time t with respect to the reference point y i Measured error data.

[0028] Preferably, the method of completing the precise positioning of the intelligent robot during operation by iterative positioning calculation based on the established intelligent robot motion model and intelligent robot reference model includes the following steps:

[0029] S231, summarizing the intelligent robot motion model and the intelligent robot reference model, and performing noise reduction by a multi-frame average noise reduction method;

[0030] Set the number of frames for multi-frame average noise reduction to U, and the multi-frame average noise reduction formula is as follows:

[0031]

[0032] in, represents the intelligent robot motion model after multi-frame average noise reduction, represents the reference model of the intelligent robot after multi-frame average noise reduction, represents the motion equation after multi-frame average denoising, Represents the reference equation after multi-frame average noise reduction;

[0033] S232. After noise reduction is completed, the precise positioning of the intelligent robot during operation is determined by positioning correction.

[0034] Calculate the distance between the intelligent robot and each reference point based on the measurement data of the reference point by the intelligent robot reference model;

[0035] The position of the intelligent robot is corrected by using a mean algorithm based on the distance between the intelligent robot and each reference point;

[0036]

[0037] in, Indicates the corrected position of the intelligent robot, represents the correction value of the i-th reference point, ΔDL i represents the distance change between the intelligent robot and the reference point, and j represents the number of reference points;

[0038] The corrected intelligent robot position is set as the precise positioning of the intelligent robot.

[0039] Preferably, the process of establishing a real-time prediction digital twin model based on precise positioning of the intelligent robot, setting a coordinate system and reference points through a data simulation algorithm process includes the following steps:

[0040] The motion model of the intelligent robot is set to be expressed as a probability function P(x t |x t-1 ,c t-1,i ), the intelligent robot reference model is expressed as a probability function P(c t,i |x t ,y i );

[0041] Calculate the one-step motion probability model of the intelligent robot through digital twins;

[0042] P(y i ,x t |c t,i ,c t-1,i )=∫P(x t |x t ,c t-1,i )P(y i ,x t-1|c t-1,i ,c t-2,i );

[0043] Calculate multi-step motion probabilities based on the calculated one-step motion probability model;

[0044] A real-time prediction digital twin model is established based on multi-step motion probability and the precise positioning of the current intelligent robot.

[0045] Preferably, the method of performing path planning on the constructed inspection network by using a path planning algorithm and inputting the path planning results into a real-time prediction digital twin model includes the following steps:

[0046] S41, initializing the population of all paths in the constructed inspection network;

[0047] The population after aggregation and initialization is set as the solution space; each intelligent robot is set as a "whale"; and the inspection target is set as the "prey";

[0048] Set the position of each whale in the solution space to: D = {D1, D2, ..., D k};

[0049] Each whale randomly chooses to surround its prey or drive it away with a bubble net, with equal probability for each whale to choose these two behaviors.

[0050] S42, after the population is initialized, the prey is tracked based on the search algorithm;

[0051] S43. Use bubble nets to drive away prey during tracking;

[0052] When whales use bubble nets to drive prey, they also constantly update their own positions;

[0053] When using the bubble net, the whale's position update formula is as follows:

[0054]

[0055] Among them, p is a constant with a default value of 1, q is a random number uniformly distributed in [-1,1], and e represents a natural constant;

[0056] S44. Iterate the whale optimization algorithm until each patrol target is completed, and output the patrol target corresponding to each intelligent robot;

[0057] The patrol target corresponding to each intelligent robot is set as the output path planning result.

[0058] Preferably, after the population is initialized, tracking prey based on a search algorithm includes the following steps:

[0059] When whales surround their prey, they will choose to swim towards the whale in the best position or towards a random whale;

[0060] When the whale swims towards the optimal position, the whale's position update formula is as follows:

[0061]

[0062] in, represents the optimal whale position at time t, A is a random number uniformly distributed in (-α, α), the initial value of α is 2 and decreases linearly with the number of iterations, and B is a random number uniformly distributed in (0, 2); represents the position of the kth whale at time t+1, represents the position of the kth whale at time t, cos(σ) represents the roulette wheel selection function, and σ represents a random angle between 0 and 360 degrees;

[0063] When swimming towards the position of a random whale, the whale's position update formula is as follows:

[0064]

[0065] in, represents the position of a random whale at time t, represents the position of the hth whale at time t+1, represents the position of the hth whale at time t;

[0066] It is assumed that when |A<1|, the whale chooses to swim towards the optimal whale position, and when |A≥1|, the whale chooses to swim towards the position of a random whale.

[0067] Preferably, the controlling of the intelligent robot based on the established real-time prediction digital twin model, the path planning result and the current precise positioning of the intelligent robot comprises the following steps:

[0068] Collect the precise positioning and energy data of the current intelligent robot in real time;

[0069] Through the control variable method, the energy consumption of the intelligent robot per unit distance is recorded. The intelligent robot energy data, the energy consumption of the intelligent robot per unit distance, the current intelligent robot's precise positioning, and the corresponding intelligent robot's path planning results are input into the real-time prediction digital twin model. The probability of the intelligent robot's inspection completion is calculated based on the real-time prediction digital twin model.

[0070] When it is detected that an intelligent robot cannot complete the inspection, the route is replanned through the path planning algorithm, and the intelligent robot is controlled based on the replanning result;

[0071] When the intelligent robot detects a fault during the inspection process, it will report it immediately and arrange for professional personnel to handle it.

[0072] The present invention also discloses a smart power plant joint inspection system based on intelligent robots, which is used to implement a smart power plant joint inspection method based on intelligent robots. The system includes: a data acquisition module, an inspection network construction module, a precise positioning module, a digital twin prediction module, and a real-time control module;

[0073] The data acquisition module is used to collect path information within the smart power plant area in real time;

[0074] The inspection network construction module is used to construct an inspection network based on the path information collected in real time, and set inspection routes and reference points;

[0075] The precise positioning module is used to precisely position the intelligent robot based on the constructed inspection network and the set reference points;

[0076] The digital twin prediction module is used to build a digital twin prediction model based on the precise positioning of the intelligent robot, and predict the movement of the intelligent robot based on the built digital twin prediction;

[0077] The real-time control module is used to control the intelligent robot based on the established real-time prediction digital twin model, path planning results and the current precise positioning of the intelligent robot.

[0078] The beneficial effects of the present invention are:

[0079] (1) The present invention collects path information within the smart power plant area in real time, and constructs an inspection network based on the collected path information. At the same time, based on the constructed inspection network, a coordinate system and reference points are set, and based on the set coordinate system and reference points, the intelligent robot in operation is accurately positioned in real time; after the positioning is completed, a real-time prediction digital twin model is established based on the precise positioning of the intelligent robot, the setting of the coordinate system and reference points through a data simulation algorithm process, and at the same time, the constructed inspection network is path planned through a path planning algorithm, and the path planning results are input into the real-time prediction digital twin model. Finally, the intelligent robot is controlled based on the established real-time prediction digital twin model, the path planning results and the current precise positioning of the intelligent robot, thereby improving the efficiency of power plant inspection.

[0080] (2) The present invention constructs an inspection network and sets a coordinate system and reference points. At the same time, based on the constructed inspection network, an intelligent robot motion model and an intelligent robot reference model are established, and the intelligent robot is accurately positioned based on the established models, thereby improving the accuracy of the intelligent robot inspection.

[0081] (3) The present invention improves the rationality of intelligent robot inspection by establishing a real-time prediction digital twin model based on the precise positioning of the intelligent robot, setting the coordinate system and reference points through a data simulation algorithm process, and calculating the movement probability of the intelligent robot by establishing a real-time prediction digital twin model.

[0082] (4) The present invention performs path planning on the constructed inspection network through a path planning algorithm, and inputs the path planning results into a real-time prediction digital twin model. At the same time, the intelligent robot is controlled based on the digital twin model, thereby improving the reliability of the intelligent robot inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0084] Figure 1 This is a flow chart of the intelligent power plant joint inspection method using an intelligent robot according to the present invention. DETAILED DESCRIPTION

[0085] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0086] In a specific embodiment of the present invention,

[0087] Reference Figure 1 As shown, the present invention provides a smart power plant joint inspection method based on intelligent robots, comprising the following steps:

[0088] S1. Collect path information within the smart power plant area in real time and build an inspection network based on the collected path information;

[0089] S2. Set the coordinate system and reference points based on the established inspection network, and accurately position the intelligent robot in real time during operation based on the set coordinate system and reference points;

[0090] S21. Setting a coordinate system and reference points based on the established inspection network;

[0091] S22. Establish an intelligent robot motion model and an intelligent robot reference model based on the constructed inspection network;

[0092] S23, based on the established intelligent robot motion model and intelligent robot reference model, through an iterative positioning calculation method, completing the precise positioning of the intelligent robot during operation;

[0093] S3. Based on the precise positioning of the intelligent robot, setting of coordinate systems and reference points, a real-time prediction digital twin model is established through a data simulation algorithm process;

[0094] S4. Path planning is performed on the constructed inspection network using a path planning algorithm, and the path planning results are input into the real-time prediction digital twin model;

[0095] S5. Control the intelligent robot based on the established real-time prediction digital twin model, path planning results, and the current precise positioning of the intelligent robot;

[0096] Further, refer to Figure 1 As shown in the figure, real-time collection of path information within the smart power plant area and construction of an inspection network based on the collected path information include the following steps:

[0097] Set up smart power plant areas and collect path information within the smart power plant areas in real time;

[0098] Set path information m By a triple {a m ,z m ,d m}composition, a m Indicates the starting point of the path information, z m Indicates the end point of the path information, d m Indicates the distance of the path;

[0099] Summarize the collected path information to build an inspection network;

[0100] Further, refer to Figure 1 As shown, setting the coordinate system and reference points based on the constructed inspection network includes the following steps:

[0101] The center of the smart power plant area is set as the coordinate origin, and the horizontal direction of the smart power plant area is set as the X-axis of the three-dimensional coordinate system, the longitudinal direction is set as the Y-axis of the three-dimensional coordinate system, and the vertical direction is set as the Z-axis of the three-dimensional coordinate system; at the same time, based on the constructed three-dimensional coordinate system, m, n, and b reference points are evenly set in the directions of the three axes at unit distances;

[0102] Further, refer to Figure 1 As shown, establishing the intelligent robot motion model and the intelligent robot reference model based on the constructed inspection network includes the following steps:

[0103] The motion model of the intelligent robot is as follows:

[0104] x t =f(x t-1 ,c t-1,i )+θ t ;

[0105] Among them, x t Indicates the position of the intelligent robot in the set coordinate system at time t, x t-1 represents the position of the intelligent robot in the set coordinate system at time t-1, c t-1,i Indicates that the intelligent robot monitoring device is at x at time t-1 t-1 Position relative to reference point y i The measurement data of f(x t-1 ,c t-1,i ) represents the position x t-1 and measurement data c t-1,i The motion equation of the intelligent robot, θ t represents the motion error of the intelligent robot at time t, which is the error distance between the set position and the actual position;

[0106] The reference model of the intelligent robot is as follows:

[0107] c t,i =g(x t ,y i )+θ t,i ;

[0108] Among them, c t,i Indicates that the intelligent robot monitoring device is at x at time t t Position relative to reference point y i The measurement data, y i Indicates the i-th reference point in the set coordinate system, g(x t ,y i ) represents the position x t and reference point y i The reference equation of the intelligent robot, θ t,i Represents the time t with respect to the reference point y i Measured error data;

[0109] Further, refer to Figure 1 As shown, based on the established intelligent robot motion model and intelligent robot reference model, the iterative positioning calculation method is used to complete the precise positioning of the intelligent robot during operation, including the following steps:

[0110] S231, summarizing the intelligent robot motion model and the intelligent robot reference model, and performing noise reduction by a multi-frame average noise reduction method;

[0111] Set the number of frames for multi-frame average noise reduction to U, and the multi-frame average noise reduction formula is as follows:

[0112]

[0113] in, represents the intelligent robot motion model after multi-frame average noise reduction, represents the reference model of the intelligent robot after multi-frame average noise reduction, represents the motion equation after multi-frame average denoising, Represents the reference equation after multi-frame average noise reduction;

[0114] S232. After noise reduction is completed, the precise positioning of the intelligent robot during operation is determined by positioning correction.

[0115] Calculate the distance between the intelligent robot and each reference point based on the measurement data of the reference point by the intelligent robot reference model;

[0116] Furthermore, the position of the intelligent robot is corrected by using a mean algorithm based on the distance between the intelligent robot and each reference point;

[0117]

[0118] in, Indicates the corrected position of the intelligent robot, Indicates the correction value of the i-th reference point, ΔDL i represents the distance change between the intelligent robot and the reference point, and j represents the number of reference points;

[0119] The corrected position of the intelligent robot is set as the precise positioning of the intelligent robot;

[0120] Further, refer to Figure 1 As shown, based on the precise positioning of the intelligent robot, setting of the coordinate system and reference points, a real-time prediction digital twin model is established through the data simulation algorithm process;

[0121] The motion model of the intelligent robot is set to be expressed as a probability function P(x t |x t-1 ,c t-1,i ), the intelligent robot reference model is expressed as a probability function P(c t,i |x t ,y i );

[0122] Calculate the one-step motion probability model of the intelligent robot through digital twins;

[0123] P(y i ,x t |ct,i ,c t-1,i )=∫P(x t |x t ,c t-1,i )P(y i ,x t-1 |c t-1,i ,c t-2,i );

[0124] Further, calculating the multi-step motion probability based on the calculated one-step motion probability model;

[0125] Furthermore, a real-time prediction digital twin model is established based on the multi-step motion probability and the precise positioning of the current intelligent robot;

[0126] Further, refer to Figure 1 As shown in the figure, the path planning algorithm is used to plan the path of the constructed inspection network, and the path planning results are input into the real-time prediction digital twin model, which includes the following steps:

[0127] S41, initializing the population of all paths in the constructed inspection network;

[0128] The population after aggregation and initialization is set as the solution space; each intelligent robot is set as a "whale"; and the inspection target is set as the "prey";

[0129] Set the position of each whale in the solution space to: D = {D1, D2, ..., D k};

[0130] Each whale randomly chooses to surround its prey or drive it away with a bubble net, with equal probability for each whale to choose these two behaviors.

[0131] S42, after the population is initialized, the prey is tracked based on the search algorithm;

[0132] When whales surround their prey, they will choose to swim towards the whale in the best position or towards a random whale;

[0133] When the whale swims towards the optimal position, the whale's position update formula is as follows:

[0134]

[0135] in, represents the optimal whale position at time t, A is a random number uniformly distributed in (-α, α), the initial value of α is 2 and decreases linearly with the number of iterations, and B is a random number uniformly distributed in (0, 2); represents the position of the kth whale at time t+1, represents the position of the kth whale at time t, cos(σ) represents the roulette wheel selection function, and σ represents a random angle between 0 and 360 degrees;

[0136] When swimming towards the position of a random whale, the whale's position update formula is as follows:

[0137]

[0138] in, represents the position of a random whale at time t, represents the position of the hth whale at time t+1, represents the position of the hth whale at time t;

[0139] Assume that when |A<1|, the whale chooses to swim towards the optimal whale position, and when |A≥1|, the whale chooses to swim towards the position of a random whale;

[0140] S43. Use bubble nets to drive away prey during tracking;

[0141] When whales use bubble nets to drive prey, they also constantly update their own positions;

[0142] When using the bubble net, the whale's position update formula is as follows:

[0143]

[0144] Among them, p is a constant with a default value of 1, q is a random number uniformly distributed in [-1,1], and e represents a natural constant;

[0145] S44. Iterate the whale optimization algorithm until each patrol target is completed, and output the patrol target corresponding to each intelligent robot;

[0146] The patrol target corresponding to each intelligent robot is set as the output path planning result;

[0147] Further, refer to Figure 1 As shown in the figure, controlling the intelligent robot based on the established real-time prediction digital twin model, path planning results, and the current precise positioning of the intelligent robot includes the following steps:

[0148] Collect the precise positioning and energy data of the current intelligent robot in real time;

[0149] Through the control variable method, the energy consumption of the intelligent robot per unit distance is recorded. The intelligent robot energy data, the energy consumption of the intelligent robot per unit distance, the current intelligent robot's precise positioning, and the corresponding intelligent robot's path planning results are input into the real-time prediction digital twin model. The probability of the intelligent robot's inspection completion is calculated based on the real-time prediction digital twin model.

[0150] When it is detected that an intelligent robot cannot complete the inspection, the route is replanned through the path planning algorithm, and the intelligent robot is controlled based on the replanning result;

[0151] Furthermore, when the intelligent robot detects a fault during the inspection process, it will immediately report it and arrange for professional personnel to handle it;

[0152] In a specific embodiment, the intelligent robot-based smart power plant joint inspection system is used to implement an intelligent robot-based smart power plant joint inspection method, and the system includes: a data acquisition module, an inspection network construction module, a precise positioning module, a digital twin prediction module, and a real-time control module;

[0153] The data acquisition module is used to collect path information within the smart power plant area in real time;

[0154] The inspection network construction module is used to construct an inspection network based on the path information collected in real time, and set inspection routes and reference points;

[0155] The precise positioning module is used to precisely position the intelligent robot based on the constructed inspection network and the set reference points;

[0156] The digital twin prediction module is used to build a digital twin prediction model based on the precise positioning of the intelligent robot, and predict the movement of the intelligent robot based on the built digital twin prediction;

[0157] The real-time control module is used to control the intelligent robot based on the established real-time prediction digital twin model, path planning results and the current precise positioning of the intelligent robot.

[0158] It should be noted that

[0159] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the scope of protection of the present invention.

Claims

1. A smart power plant joint inspection method based on intelligent robots is characterized by: The following steps are involved: S1. Collect path information within the smart power plant area in real time and build an inspection network based on the collected path information; S2. Set the coordinate system and reference points based on the established inspection network, and accurately position the intelligent robot in real time based on the set coordinate system and reference points; S21. Setting a coordinate system and reference points based on the established inspection network; S22. Establish an intelligent robot motion model and an intelligent robot reference model based on the constructed inspection network; S23, based on the established intelligent robot motion model and intelligent robot reference model, through an iterative positioning calculation method, completing the precise positioning of the intelligent robot during operation; S3. Based on the precise positioning of the intelligent robot, setting of coordinate systems and reference points, a real-time prediction digital twin model is established through a data simulation algorithm process; S4. Path planning is performed on the constructed inspection network using a path planning algorithm, and the path planning results are input into the real-time prediction digital twin model; S5. Control the intelligent robot based on the established real-time prediction digital twin model, path planning results and the current precise positioning of the intelligent robot.

2. The intelligent power plant joint inspection method based on intelligent robots according to claim 1 is characterized in that: The real-time collection of path information within the smart power plant area and the construction of an inspection network based on the collected path information include the following steps: Set up smart power plant areas and collect path information within the smart power plant areas in real time; Set path information m By a triple {a m ,z m ,d m }composition, a m Indicates the starting point of the path information, z m Indicates the end point of the path information, d m Indicates the distance of the path; The collected path information is aggregated to build an inspection network.

3. The intelligent power plant joint inspection method based on intelligent robots according to claim 1 is characterized in that: The setting of the coordinate system and reference points based on the constructed inspection network includes the following steps: The center area of ​​the smart power plant is set as the coordinate origin, and the horizontal direction of the smart power plant area is used as the X-axis of the three-dimensional coordinate system, the longitudinal direction is used as the Y-axis of the three-dimensional coordinate system, and the vertical direction is used as the Z-axis of the three-dimensional coordinate system; at the same time, based on the constructed three-dimensional coordinate system, m, n, and b reference points are evenly set in the directions of the three axes according to unit distances.

4. The intelligent power plant joint inspection method based on intelligent robots according to claim 1 is characterized in that: The intelligent robot motion model and intelligent robot reference model are established based on the constructed inspection network The following steps are involved: The motion model of the intelligent robot is as follows: x t =f(x t-1 ,c t-1,i )+θ t ; Among them, x t Indicates the position of the intelligent robot in the set coordinate system at time t, x t-1 represents the position of the intelligent robot in the set coordinate system at time t-1, c t-1,i Indicates that the intelligent robot monitoring device is at x at time t-1 t-1 Position relative to reference point y i The measurement data of f(x t-1 ,c t-1,i ) represents the position x t-1 and measurement data c t-1,i The motion equation of the intelligent robot, θ t represents the motion error of the intelligent robot at time t, which is the error distance between the set position and the actual position; The reference model of the intelligent robot is as follows: c t,i =g(x t ,y i )+θ t,i 4 Among them, c t,i Indicates that the intelligent robot monitoring device is at x at time t t Position relative to reference point y i The measurement data, y i Indicates the i-th reference point in the set coordinate system, g(x t ,y i ) represents the position x t and reference point y i The reference equation of the intelligent robot, θ t,i Represents the time t with respect to the reference point y i Measured error data.

5. The intelligent power plant joint inspection method based on intelligent robots according to claim 1 is characterized in that: The method of accurately positioning the intelligent robot during operation by iterative positioning calculation based on the established intelligent robot motion model and intelligent robot reference model includes the following steps: S231, summarizing the intelligent robot motion model and the intelligent robot reference model, and performing noise reduction by a multi-frame average noise reduction method; Set the number of frames for multi-frame average noise reduction to U, and the multi-frame average noise reduction formula is as follows: in, represents the intelligent robot motion model after multi-frame average noise reduction, represents the reference model of the intelligent robot after multi-frame average noise reduction, represents the motion equation after multi-frame average denoising, Represents the reference equation after multi-frame average noise reduction; S232. After noise reduction is completed, the precise positioning of the intelligent robot during operation is determined by positioning correction. Calculate the distance between the intelligent robot and each reference point based on the measurement data of the reference point by the intelligent robot reference model; The position of the intelligent robot is corrected by using a mean algorithm based on the distance between the intelligent robot and each reference point; in, Indicates the corrected position of the intelligent robot, represents the correction value of the i-th reference point, ΔDL i represents the distance change between the intelligent robot and the reference point, and j represents the number of reference points; The corrected intelligent robot position is set as the precise positioning of the intelligent robot.

6. The intelligent power plant joint inspection method based on intelligent robots according to claim 1 is characterized in that: The process of establishing a real-time prediction digital twin model based on the precise positioning of the intelligent robot, setting the coordinate system and reference points through a data simulation algorithm includes the following steps: The motion model of the intelligent robot is set to be expressed as a probability function P(x t |x t-1 ,c t-1,i ), the intelligent robot reference model is expressed as a probability function P(c t,i |x t ,y i ); Calculate the one-step motion probability model of the intelligent robot through digital twins; P(y i ,x t |c t,i ,c t-1,i )=∫P(x t |x t ,c t-1,i )P(y i ,x t-1 |c t-1,i ,c t-2,i ); Calculate multi-step motion probabilities based on the calculated one-step motion probability model; A real-time prediction digital twin model is established based on multi-step motion probability and the precise positioning of the current intelligent robot.

7. The intelligent power plant joint inspection method based on intelligent robots according to claim 1 is characterized in that: The method of performing path planning on the constructed inspection network by using a path planning algorithm and inputting the path planning results into a real-time prediction digital twin model includes the following steps: S41, initializing the population of all paths in the constructed inspection network; Set the population after aggregation and initialization as the solution space; set each intelligent robot as a "whale"; set the inspection target as "prey"; Set the position of each whale in the solution space to: D = {D1, D2, ..., D k }; Each whale randomly chooses to surround its prey or drive it away with a bubble net, with equal probability for each whale to choose these two behaviors. S42, after the population is initialized, the prey is tracked based on the search algorithm; S43. Use bubble nets to drive away prey during tracking; When whales use bubble nets to drive prey, they also constantly update their own positions; When using the bubble net, the whale's position update formula is as follows: Among them, p is a constant with a default value of 1, q is a random number uniformly distributed in [-1,1], and e represents a natural constant; S44. Iterate the whale optimization algorithm until each patrol target is completed, and output the patrol target corresponding to each intelligent robot; The patrol target corresponding to each intelligent robot is set as the output path planning result.

8. The intelligent robot-based smart power plant joint inspection method according to claim 1 is characterized in that: After the population is initialized, tracking prey based on the search algorithm includes the following steps: When whales surround their prey, they will choose to swim towards the whale in the best position or towards a random whale; When the whale swims towards the optimal position, the whale's position update formula is as follows: in, represents the optimal whale position at time t, A is a random number uniformly distributed in (-α, α), the initial value of α is 2 and decreases linearly with the number of iterations, and B is a random number uniformly distributed in (0, 2); represents the position of the kth whale at time t+1, represents the position of the kth whale at time t, cos(σ) represents the roulette wheel selection function, and σ represents a random angle between 0 and 360 degrees; When swimming towards the position of a random whale, the whale's position update formula is as follows: in, represents the position of a random whale at time t, represents the position of the hth whale at time t+1, represents the position of the hth whale at time t; It is assumed that when |A<1|, the whale chooses to swim towards the optimal whale position, and when |A≥1|, the whale chooses to swim towards the position of a random whale.

9. The intelligent power plant joint inspection method based on intelligent robots according to claim 1 is characterized in that: The control of the intelligent robot based on the established real-time prediction digital twin model, the path planning results and the current precise positioning of the intelligent robot includes the following steps: Collect the precise positioning and energy data of the current intelligent robot in real time; Through the control variable method, the energy consumption of the intelligent robot per unit distance is recorded. The intelligent robot energy data, the energy consumption of the intelligent robot per unit distance, the current intelligent robot's precise positioning, and the corresponding intelligent robot's path planning results are input into the real-time prediction digital twin model. The probability of the intelligent robot's inspection completion is calculated based on the real-time prediction digital twin model. When it is detected that an intelligent robot cannot complete the inspection, the route is replanned through the path planning algorithm, and the intelligent robot is controlled based on the replanning result; When the intelligent robot detects a fault during the inspection process, it will report it immediately and arrange for professional personnel to handle it.

10. A system for implementing the intelligent robot-based smart power plant joint inspection method according to any one of claims 1 to 9, characterized in that: include: Data acquisition module, inspection network construction module, precise positioning module, digital twin prediction module and real-time control module; The data acquisition module is used to collect path information within the smart power plant area in real time; The inspection network construction module is used to construct an inspection network based on the path information collected in real time, and set inspection routes and reference points; The precise positioning module is used to precisely position the intelligent robot based on the constructed inspection network and the set reference points; The digital twin prediction module is used to build a digital twin prediction model based on the precise positioning of the intelligent robot, and predict the movement of the intelligent robot based on the built digital twin prediction; The real-time control module is used to control the intelligent robot based on the established real-time prediction digital twin model, path planning results and the current precise positioning of the intelligent robot.

Images