A dynamic deployment method and device for AI robots

Abstract

The present invention discloses a dynamic deployment method and device for AI robots, which relate to the technical field of dynamic deployment of robots, and obtain the volume information of all industrial equipment, the information of measuring instruments deployed on the industrial equipment, and the distance information between the industrial equipment; determine the first information acquisition rate of each industrial equipment; determine the information increment of each industrial equipment obtained when the AI ​​robot monitors the industrial equipment; determine the monitoring order of the industrial equipment by the AI ​​robot based on the information increment of each industrial equipment; the AI ​​robot monitors the industrial equipment according to the determined monitoring order; deploy the AI ​​robot, and adjust the working order of the robot according to the operating status of the equipment to improve the working efficiency and accuracy of the robot; comprehensively obtain the information of the industrial equipment through the AI ​​robot, solve the problem that the measuring instrument can only obtain local information of the industrial equipment, and improve the monitoring effect of the industrial equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic deployment of robots, and in particular to a dynamic deployment method and device applied to AI robots. Background Art

[0002] In the industrial production process, dynamic deployment of robots is an advanced production management technology that can flexibly deploy and adjust robots according to production needs and real-time environmental changes, thereby improving production efficiency and the monitoring effect of industrial equipment. At present, in the process of monitoring industrial equipment, the parameter information of industrial equipment is usually obtained by using measuring instruments deployed on the industrial equipment. In this case, only the parameter information of part of the industrial equipment can be obtained, and the parameter information of the part of the industrial equipment where the measuring instruments are not deployed is not obtained. When an abnormal situation occurs in the part of the industrial equipment where the measuring instruments are not deployed, it is difficult to discover it in time; therefore, how to obtain more parameter information of industrial equipment through dynamic deployment of robots, and then improve the monitoring effect of industrial equipment, has become an urgent problem to be solved. Summary of the invention

[0003] The purpose of the present invention is to provide a dynamic deployment method and device for AI robots to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solution: a dynamic deployment method applied to an AI robot, comprising:

[0005] S11, obtaining volume information of all industrial equipment, information of measuring instruments deployed on the industrial equipment, and distance information between the industrial equipment;

[0006] S12, determining a first information acquisition rate of each industrial device according to the measuring instrument information deployed on each industrial device, where the first information acquisition rate is the proportion of device information acquired only through the measuring instrument to all device information;

[0007] Since measuring instruments can only obtain partial parameter information of industrial equipment, the rest of the parameter information of industrial equipment needs to be obtained by deploying AI robots to the industrial equipment. The first information acquisition rate of industrial equipment is related to the volume of the industrial equipment and the parameter variables involved. The larger the volume of the industrial equipment, the more data the measuring instrument loses. An industrial equipment usually needs to measure more than one parameter, so the measuring instrument is also related to the number of parameter variables that need to be measured by the industrial equipment.

[0008] S13, based on the first information acquisition rate of each industrial device, determining the information increment of each industrial device obtained when the AI ​​robot monitors the industrial device; and determining the monitoring order of the industrial device by the AI ​​robot based on the information increment of each industrial device;

[0009] S14, the AI ​​robot monitors the industrial equipment according to the determined monitoring sequence and obtains the industrial equipment information that the measuring instruments cannot obtain.

[0010] In step S12, the first information acquisition rate of each industrial device is calculated by the following formula:

[0011] H i =f i (V igc ) / f i (V i ), where H i represents the first information acquisition rate of the i-th industrial equipment, V i represents the volume of the i-th industrial equipment, V igc represents the total volume of the area where the i-th industrial equipment can be detected by the measuring instrument; f i (V i ) and the number of variables and volume V that need to be detected by the i-th industrial equipment i Proportional to; f i (V igc ) and the number of variables that need to be detected by the i-th industrial equipment and the total volume V of the area that can be detected by the measuring instrument of the i-th industrial equipment igc Directly proportional.

[0012] The total volume V of the area where the i-th industrial equipment can be detected by the measuring instrument igc This is determined by the following steps:

[0013] On the i-th industrial equipment, determine the positions of all measuring instruments; for the k-th measuring instrument on the i-th industrial equipment, actively apply disturbance to the i-th industrial equipment in the surrounding area of ​​the k-th measuring instrument. If the k-th measuring instrument on the i-th industrial equipment detects that the parameters of the industrial equipment have changed within a time period, then it is judged that the k-th measuring instrument can detect the information of the i-th industrial equipment in the surrounding area; if the k-th measuring instrument on the i-th industrial equipment fails to detect the parameter changes of the industrial equipment within the time period, then it is judged that the k-th measuring instrument cannot detect the information of the i-th industrial equipment in the surrounding area; determine all surrounding areas of the k-th measuring instrument that can detect the information of the i-th industrial equipment in the order of the distance from the k-th measuring instrument on the i-th industrial equipment from small to large, and merge to obtain the area of ​​the industrial equipment that the k-th measuring instrument can detect; merge the areas of the industrial equipment that all measuring instruments can detect to obtain the total volume V of the area where the i-th industrial equipment can be detected by the measuring instrument igc .

[0014] In step S13, the information increment of each industrial device is determined by the following formula:

[0015] Ipr i =(1-H i )×f i (V i ),Ipr i Where represents the information increment of the i-th industrial equipment.

[0016] When AI robots are deployed at industrial equipment, they can obtain comprehensive parameter information of the industrial equipment, while measuring instruments can only obtain local information of the industrial equipment, thus generating incremental information.

[0017] In step S13, determining the monitoring order of the industrial equipment by the AI ​​robot based on the information increment of each industrial equipment further includes the following steps:

[0018] S51, select the initial temperature T0, randomly select industrial equipment to obtain the monitoring order of the industrial equipment, the same industrial equipment is allowed to exist in the monitoring order, and the order in which the industrial equipment is monitored by the AI ​​robot is allowed to be more than once; the industrial equipment monitoring order is selected as the initial solution, and the initial solution is set as the current solution to determine whether the current solution meets the constraint of the AI ​​robot monitoring time. If the constraint of the AI ​​robot monitoring time is met, determine the estimated value freq0 corresponding to the current solution, and set the initial temperature to the current temperature; if the constraint of the AI ​​robot monitoring time is not met, randomly select industrial equipment again to obtain the monitoring order of the industrial equipment;

[0019] S52, set the counting variable num, and repeat steps S53 to S54 for num=1, 2, ..., N; N represents the number of times of the set inner loop;

[0020] S53, by generating disturbances based on the current solution, changing the monitoring order of the industrial equipment, and determining whether the monitoring order of the industrial equipment after the disturbances are generated meets the constraint conditions of the AI ​​robot monitoring time; if the constraint conditions are met, the monitoring order of the industrial equipment after the disturbances is used as a new solution, and the sub-evaluation freq corresponding to the new solution is determined; if not, the disturbances are re-generated based on the current solution and the monitoring order of the industrial equipment is changed;

[0021] S54, calculating the increment Δfreq of the sub-estimation value brought by the new solution, if the increment of the sub-estimation value is less than 0, accepting the new solution as the new current solution with probability 1, if the increment of the sub-estimation value is greater than or equal to 0, accepting the new solution as the new current solution with probability exp(-Δfreq / T), where T represents the current temperature;

[0022] S55, after completing one iteration, select a cooling solution and lower the current temperature. If the current temperature is not less than the temperature threshold, enter step S52; if the current temperature is less than the threshold, determine the monitoring order of the industrial equipment according to the current solution.

[0023] In step S51, the following steps are used to determine whether the current solution meets the constraints of the AI ​​robot monitoring time:

[0024] S61, obtaining the time required for the AI ​​robot to move between two different industrial devices and the time required for the AI ​​robot to complete one monitoring of each industrial device;

[0025] S62, add the time required for the AI ​​robot to complete one monitoring of the industrial equipment included in the current solution to obtain the monitoring time of the current solution; add the time consumed by the AI ​​robot when moving according to the industrial equipment monitoring sequence corresponding to the current solution to obtain the movement time of the current solution; add the movement time and monitoring time of the current solution to obtain the time constraint of the current solution. If the time constraint of the current solution is not greater than the monitoring time Tset set by the AI ​​robot, the current solution meets the constraint of the AI ​​robot's monitoring time. Otherwise, the current solution does not meet the constraint of the AI ​​robot's monitoring time.

[0026] Determining the industrial equipment causing the disturbance also includes the following steps:

[0027] Let the monitoring order of industrial equipment corresponding to the current solution be Ax1, Ax2, ..., Axm, where Ax1, Ax2, ..., Axm represent industrial equipment and m is the number of industrial equipment monitored by the AI ​​robot; Ax1, Ax2, ..., Axm allow the existence of the same industrial equipment;

[0028] Let the information increments of Ax1, Ax2, ..., Axm be Ipr respectively x1 , Ipr x2 ,…,Ipr xm , calculate the inverse of the information increment 1 / Ipr x1 , 1 / Ipr x2 ,…,1 / Ipr xm , the sum of the reciprocals of the information increment is taken as the length of the random interval, and m random subintervals are obtained according to the reciprocals of the information increment. The random subintervals are [0, Ipr x1 ]、[1 / Ipr x1 ,1 / Ipr x1 +1 / Ipr x2 ], …, [1 / Ipr x1 +…+1 / Ipr xm-1 ,1 / Ipr x1 +…+1 / Ipr xm ], the subintervals correspond one-to-one to the industrial equipment Ax1, Ax2, …, Axm;

[0029] All random subintervals are concatenated into a random interval, and in the random interval [0,1 / Ipr x1 +…+1 / Ipr xm ]Generate a random number and find the industrial equipment corresponding to the random sub-interval where the random number is located. The industrial equipment will be replaced by a new industrial equipment; when the random number is at the end point of the random sub-interval, regenerate a random number to determine the replaced industrial equipment.

[0030] The more information increments of industrial equipment, the more additional industrial equipment information obtained by AI robots, the more helpful it is for monitoring industrial equipment. Therefore, the industrial equipment that has been disturbed is determined based on the information increments of industrial equipment.

[0031] Generate disturbances to industrial equipment by following these steps:

[0032] S81, let the replaced industrial equipment be Axq, where q is a positive integer between the interval [1, m];

[0033] S82, under the monitoring sequence of the industrial equipment corresponding to the current solution, analyze the conditional information increment when the industrial equipment replaces the industrial equipment Axq at the position of the industrial equipment Axq, and let TIpr i represents the condition information increment of industrial equipment Ai, t i Indicates the time since the industrial equipment Ai was last monitored by the AI ​​robot. If t i If TIpr is less than the safety time of industrial equipment Ai, i =0; if t iNot less than the safety time of industrial equipment Ai, then TIpr i =Ipr i ×t i The safety time of the industrial equipment Ai refers to the time during which the parameter information of the industrial equipment Ai is consistent with the parameter information when the AI ​​robot monitors the industrial equipment Ai after the industrial equipment Ai is monitored by the AI ​​robot;

[0034] S83, concatenate the condition information increments of the industrial equipment except Axq into an interval, and obtain M sub-intervals, where M+1 is the number of industrial equipment, and the sub-intervals are [0, TIpr1], [TIpr1, TIpr1+TIpr2], …, [TIpr1+…+TIpr M-1 ,TIpr1+…+TIpr M ], the sub-intervals correspond to the production information one by one; in the interval [0,TIpr1+…+TIpr M ]Generate a random number and determine the industrial equipment corresponding to the sub-interval where the random number is located. The industrial equipment will replace Axq to form a new solution; when the random number is at the endpoint of the sub-interval, a new random number is generated.

[0035] When the AI ​​robot completes a monitoring of industrial equipment, combined with the measurement results of the measuring instrument, it can be determined that the probability of abnormal conditions occurring in the industrial equipment in the next period of time is low. At this time, it is not very meaningful to continue to use the AI ​​robot to monitor the industrial equipment and obtain comprehensive information about the industrial equipment. Therefore, deploying AI robots to monitor the same industrial equipment in a short period of time can only generate one effective information increment, and the second conditional information increment is zero. When industrial equipment has not been fully monitored by the AI ​​robot for a long time, there may be abnormal conditions that have not been detected by the measuring instrument. Therefore, the conditional information increment is positively correlated with the time since the industrial equipment was last monitored by the AI ​​robot.

[0036] In steps S51 and S53, the estimated values ​​of the monitoring sequence are determined by:

[0037] Get the sum of the condition information increments ΣIpr of all industrial equipment in the monitoring sequence, calculate the condition information increment acquisition speed DV, DV = ΣIpr / Tset, and take the inverse of the condition information increment acquisition speed DV 1 / DV to get the sub-valuation value of the monitoring sequence. The duration of a single monitoring by the AI ​​robot is Tset. During this time period, it is necessary to obtain the information of industrial equipment as effectively as possible. The sub-valuation value is positively correlated with the inverse of the condition information increment acquisition speed.

[0038] To achieve the above-mentioned objectives, the present invention provides the following technical solutions: a dynamic deployment device applied to an AI robot, comprising: an AI robot module and a data analysis module; the AI ​​robot module is used to control the AI ​​robot to move to a deployment position according to a monitoring sequence, and to perform a comprehensive detection of industrial equipment parameters; the data analysis module determines the deployment position of the AI ​​robot and the order of going to the deployment position when the AI ​​robot is monitoring.

[0039] Compared with the prior art, the beneficial effects of the present invention are: deploying AI robots, adjusting the working order of the robots according to the operating status of the equipment, and improving the working efficiency and accuracy of the robots; obtaining comprehensive information on industrial equipment through AI robots, solving the problem that measuring instruments can only obtain local information on industrial equipment, and improving the monitoring effect of industrial equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 The present invention is a flow chart of a dynamic deployment method applied to an AI robot. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0042] Example: Figure 1 As shown, the present invention provides a technical solution, a dynamic deployment method applied to an AI robot, comprising:

[0043] S11, obtaining volume information of all industrial equipment, information of measuring instruments deployed on the industrial equipment, and distance information between the industrial equipment;

[0044] S12, determine the total volume V of the area where the i-th industrial equipment can be detected by the measuring instrument igc :

[0045] On the i-th industrial equipment, determine the positions of all measuring instruments; for the k-th measuring instrument on the i-th industrial equipment, actively apply disturbance to the i-th industrial equipment in the surrounding area of ​​the k-th measuring instrument. If the k-th measuring instrument on the i-th industrial equipment detects that the parameters of the industrial equipment have changed within a time period, then it is judged that the k-th measuring instrument can detect the information of the i-th industrial equipment in the surrounding area; if the k-th measuring instrument on the i-th industrial equipment fails to detect the parameter changes of the industrial equipment within the time period, then it is judged that the k-th measuring instrument cannot detect the information of the i-th industrial equipment in the surrounding area; determine all surrounding areas of the k-th measuring instrument that can detect the information of the i-th industrial equipment in the order of the distance from the k-th measuring instrument on the i-th industrial equipment from small to large, and merge to obtain the area of ​​the industrial equipment that the k-th measuring instrument can detect; merge the areas of the industrial equipment that all measuring instruments can detect to obtain the total volume V of the area where the i-th industrial equipment can be detected by the measuring instrument igc .

[0046] According to the measuring instrument information deployed on each industrial device, the first information acquisition rate of each industrial device is determined, where the first information acquisition rate is the proportion of device information acquired only through the measuring instrument in all device information; the first information acquisition rate of each industrial device is calculated by the following formula:

[0047] H i =f i (V igc ) / f i (V i ), where H i represents the first information acquisition rate of the i-th industrial equipment, V i represents the volume of the i-th industrial equipment, V igc represents the total volume of the area where the i-th industrial equipment can be detected by the measuring instrument; f i (V i ) and the number of variables and volume V that need to be detected by the i-th industrial equipment i Proportional to; f i (V igc ) and the number of variables that need to be detected by the i-th industrial equipment and the total volume V of the area that can be detected by the measuring instrument of the i-th industrial equipment igc Directly proportional.

[0048] S13, based on the first information acquisition rate of each industrial device, determining the information increment of each industrial device obtained when the AI ​​robot monitors the industrial device:

[0049] Ipr i =(1-H i )×f i(V i ),Ipr i Where represents the information increment of the i-th industrial equipment.

[0050] When the volume of the i-th industrial equipment is 50 cubic meters, the parameter variables to be measured are voltage and current, and the total volume of the area that can be detected by the measuring instrument is 10 cubic meters, optionally, a linear mapping is directly performed, f i (V i )=100,f i (V igc )=20, then the first information acquisition rate of the i-th industrial equipment is 0.2. After deploying the AI ​​robot to monitor the i-th industrial equipment, the information increment of the i-th industrial equipment is 80;

[0051] Determine the order in which AI robots monitor industrial equipment based on the information increment of each industrial equipment:

[0052] S51, select the initial temperature T0, randomly select industrial equipment to obtain the monitoring order of the industrial equipment, the same industrial equipment is allowed to exist in the monitoring order, and the order in which the industrial equipment is monitored by the AI ​​robot is allowed to be more than once; the industrial equipment monitoring order is selected as the initial solution, and the initial solution is set as the current solution to determine whether the current solution meets the constraint of the AI ​​robot monitoring time. If the constraint of the AI ​​robot monitoring time is met, determine the estimated value freq0 corresponding to the current solution, and set the initial temperature to the current temperature; if the constraint of the AI ​​robot monitoring time is not met, randomly select industrial equipment again to obtain the monitoring order of the industrial equipment;

[0053] The initial temperature T0 will affect the number of iterations, and thus the search results and calculation speed. The initial temperature T0 can be set according to the computing resources available.

[0054] Use the following steps to determine whether the current solution meets the constraints of the AI ​​robot's monitoring time:

[0055] Obtain the time required for the AI ​​robot to move between two different industrial equipment and the time required for the AI ​​robot to complete a monitoring of each industrial equipment; add the time required for the AI ​​robot to complete a monitoring of the industrial equipment included in the current solution to obtain the monitoring time of the current solution; add the time consumed by the AI ​​robot when moving according to the industrial equipment monitoring sequence corresponding to the current solution to obtain the moving time of the current solution; add the moving time and the monitoring time of the current solution to obtain the time constraint of the current solution. If the time constraint of the current solution is not greater than the monitoring time Tset set by the AI ​​robot, the current solution meets the constraint of the AI ​​robot monitoring time. Otherwise, the current solution does not meet the constraint of the AI ​​robot monitoring time.

[0056] When two identical industrial devices appear consecutively, there is no moving time, only the time required for two monitorings.

[0057] S52, set the counting variable num, and repeat steps S53 to S54 for num=1, 2, ..., N; N represents the number of times of the set inner loop;

[0058] S53, by generating disturbances based on the current solution, changing the monitoring order of the industrial equipment, and determining whether the monitoring order of the industrial equipment after the disturbances are generated meets the constraint conditions of the AI ​​robot monitoring time; if the constraint conditions are met, the monitoring order of the industrial equipment after the disturbances is used as a new solution, and the sub-evaluation freq corresponding to the new solution is determined; if not, the disturbances are re-generated based on the current solution and the monitoring order of the industrial equipment is changed;

[0059] First identify the industrial equipment where the disturbance occurs:

[0060] Let the monitoring order of industrial equipment corresponding to the current solution be Ax1, Ax2, ..., Axm, where Ax1, Ax2, ..., Axm represent industrial equipment and m is the number of industrial equipment monitored by the AI ​​robot; Ax1, Ax2, ..., Axm allow the existence of the same industrial equipment;

[0061] Let the information increments of Ax1, Ax2, ..., Axm be Ipr respectively x1 , Ipr x2 ,…,Ipr xm , calculate the inverse of the information increment 1 / Ipr x1 , 1 / Ipr x2 ,…,1 / Ipr xm , the sum of the reciprocals of the information increment is taken as the length of the random interval, and m random subintervals are obtained according to the reciprocals of the information increment. The random subintervals are [0, Ipr x1 ]、[1 / Ipr x1 ,1 / Ipr x1 +1 / Ipr x2 ], …, [1 / Ipr x1 +…+1 / Ipr xm-1 ,1 / Ipr x1 +…+1 / Ipr xm ], the subintervals correspond one-to-one to the industrial equipment Ax1, Ax2, …, Axm;

[0062] All random subintervals are concatenated into a random interval, and in the random interval [0,1 / Ipr x1 +…+1 / Ipr xm]Generate a random number and find the industrial equipment corresponding to the random sub-interval where the random number is located. The industrial equipment will be replaced by a new industrial equipment; when the random number is at the end point of the random sub-interval, regenerate a random number to determine the replaced industrial equipment.

[0063] Next, determine which industrial equipment will replace the disrupted industrial equipment:

[0064] Let the replaced industrial equipment be Axq, where q is a positive integer between the interval [1, m];

[0065] Under the monitoring order of industrial equipment corresponding to the current solution, analyze the conditional information increment when the industrial equipment replaces the industrial equipment Axq at the position of the industrial equipment Axq, and let TIpr i represents the condition information increment of industrial equipment Ai, t i Indicates the time since the industrial equipment Ai was last monitored by the AI ​​robot. If t i If TIpr is less than the safety time of industrial equipment Ai, i =0; if t i Not less than the safety time of industrial equipment Ai, then TIpr i =Ipr i ×t i The safety time of the industrial equipment Ai refers to the time during which the parameter information of the industrial equipment Ai is consistent with the parameter information when the AI ​​robot monitors the industrial equipment Ai after the industrial equipment Ai is monitored by the AI ​​robot;

[0066] The condition information increments of industrial equipment except Axq are spliced ​​into an interval, and M sub-intervals are obtained, where M+1 is the number of industrial equipment, and the sub-intervals are [0,TIpr1], [TIpr1,TIpr1+TIpr2], …, [TIpr1+…+TIpr M-1 ,TIpr1+…+TIpr M ], the sub-intervals correspond to the production information one by one; in the interval [0,TIpr1+…+TIpr M ]Generate a random number and determine the industrial equipment corresponding to the sub-interval where the random number is located. The industrial equipment will replace Axq to form a new solution; when the random number is at the endpoint of the sub-interval, a new random number is generated.

[0067] In particular, when the safety time of industrial equipment Ai is 30 minutes, there are 3 industrial equipment Ai in the monitoring sequence, and the time intervals are all 25 minutes. Then the condition information increment of the third industrial equipment Ai is not set to zero, and the condition information increment is analyzed as if there is no second industrial equipment Ai. The same applies to the calculation of the condition information increment acquisition speed DV. Broadly, if there is a situation where industrial equipment Ai is set to zero, then in the monitoring sequence, find the previous industrial equipment Ai and the next industrial equipment Ai. For the next industrial equipment Ai, when determining the condition information increment, ignore the influence of the intermediate industrial equipment Ai; the operation of ignoring the influence can be transferred. For example, when the safety time of industrial equipment Ai is 40 minutes, there are 4 industrial equipment Ai in the monitoring sequence, and the time intervals are all 15 minutes. Then first determine the condition information increment of the second industrial equipment Ai. Zero, ignore the second industrial equipment Ai, the time interval between the first and third industrial equipment Ai is 30 minutes, which is less than the safety time, continue to ignore the three industrial equipment Ai, the time interval between the first and fourth industrial equipment Ai is 45 minutes, and the conditional information increment is not zero; when the AI ​​robot monitors these four industrial equipment Ai, only the first and fourth industrial equipment Ai are necessary, and AI's monitoring of the second and third industrial equipment Ai can be reflected by monitoring the first industrial equipment Ai.

[0068] The safety time of industrial equipment Ai can be analyzed based on the historical data of the parameter variables of the industrial equipment. For example, after the AI ​​robot completes the monitoring of the industrial equipment Ai, it obtains a curve graph of the parameter variables. Based on the historical data, it knows the time when the next trend of the parameter variables is roughly consistent with the known curve graph. Whether it is consistent can be determined by calculating the mean absolute percentage error, root mean square error, etc.

[0069] Finally, calculate the estimated value of the monitoring sequence:

[0070] The sum of the condition information increments ΣIpr of all industrial equipment in the monitoring sequence is obtained, and the condition information increment acquisition speed DV is calculated, DV=ΣIpr / Tset, and the reciprocal of the condition information increment acquisition speed DV 1 / DV is taken to obtain the estimated value of the monitoring sequence.

[0071] S54, calculating the increment Δfreq of the sub-estimation value brought by the new solution, if the increment of the sub-estimation value is less than 0, accepting the new solution as the new current solution with probability 1, if the increment of the sub-estimation value is greater than or equal to 0, accepting the new solution as the new current solution with probability exp(-Δfreq / T), where T represents the current temperature;

[0072] S55, after completing one iteration, select a cooling solution and lower the current temperature. If the current temperature is not less than the temperature threshold, enter step S52; if the current temperature is less than the threshold, determine the monitoring order of the industrial equipment according to the current solution.

[0073] S14, the AI ​​robot monitors the industrial equipment according to the determined monitoring sequence and obtains the industrial equipment information that the measuring instruments cannot obtain.

[0074] Embodiment: The present invention provides a technical solution, a dynamic deployment device applied to an AI robot, comprising: an AI robot module and a data analysis module; the AI ​​robot module is used to control the AI ​​robot to move to a deployment position according to a monitoring sequence, and perform a comprehensive detection of industrial equipment parameters; the data analysis module determines the deployment position of the AI ​​robot and the order of going to the deployment position when the AI ​​robot is monitoring.

[0075] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A dynamic deployment method applied to an AI robot, characterized in that: The following steps are involved: S11, obtaining volume information of all industrial equipment, information of measuring instruments deployed on the industrial equipment, and distance information between the industrial equipment; S12, determining a first information acquisition rate of each industrial device according to the measuring instrument information deployed on each industrial device, where the first information acquisition rate is the proportion of device information acquired only through the measuring instrument to all device information; S13, based on the first information acquisition rate of each industrial device, determining the information increment of each industrial device obtained when the AI ​​robot monitors the industrial device; Determine the order in which AI robots monitor industrial equipment based on the information increment of each industrial equipment; S14, the AI ​​robot monitors the industrial equipment according to the determined monitoring sequence and obtains the industrial equipment information that the measuring instruments cannot obtain.

2. A dynamic deployment method for AI robots according to claim 1, characterized in that: In step S12, the first information acquisition rate of each industrial device is calculated by the following formula: H i =f i (V igc ) / f i (V i ), where H i represents the first information acquisition rate of the i-th industrial equipment, V i represents the volume of the i-th industrial equipment, V igc represents the total volume of the area where the i-th industrial equipment can be detected by the measuring instrument; f i (V i ) and the number of variables and volume V that need to be detected by the i-th industrial equipment i Proportional to; f i (V igc ) and the number of variables that need to be detected by the i-th industrial equipment and the total volume V of the area that can be detected by the measuring instrument of the i-th industrial equipment igc Directly proportional.

3. A dynamic deployment method for AI robots according to claim 2, characterized in that: The total volume V of the area where the i-th industrial equipment can be detected by the measuring instrument igc This is determined by the following steps: On the i-th industrial equipment, determine the locations of all measuring instruments; for the k-th measuring instrument on the i-th industrial equipment, actively apply disturbance to the i-th industrial equipment in the surrounding area of ​​the k-th measuring instrument. If the k-th measuring instrument on the i-th industrial equipment detects that the parameters of the industrial equipment have changed within a time period, then it is determined that the k-th measuring instrument can detect the information of the i-th industrial equipment in the surrounding area; If the kth measuring instrument on the i-th industrial equipment fails to detect a change in the parameters of the industrial equipment within a time period, it is determined that the kth measuring instrument cannot detect the information of the i-th industrial equipment in the surrounding area; all surrounding areas of the kth measuring instrument that can detect the information of the i-th industrial equipment are determined in ascending order of distance from the kth measuring instrument on the i-th industrial equipment, and the areas of the industrial equipment that can be detected by the kth measuring instrument are merged; the areas of the industrial equipment that can be detected by all measuring instruments are merged to obtain the total volume V of the area where the i-th industrial equipment can be detected by the measuring instrument. igc .

4. The dynamic deployment method for AI robots according to claim 3, characterized in that: In step S13, the information increment of each industrial device is determined by the following formula: Ipr i =(1-H i )×f i (V i ),Ipr i Where represents the information increment of the i-th industrial equipment.

5. A dynamic deployment method for AI robots according to claim 4, characterized in that: In step S13, determining the monitoring order of the industrial equipment by the AI ​​robot based on the information increment of each industrial equipment further includes the following steps: S51, select the initial temperature T0, randomly select industrial equipment to obtain the monitoring order of the industrial equipment, the same industrial equipment is allowed to exist in the monitoring order, and the order in which the industrial equipment is monitored by the AI ​​robot is allowed to be more than once; the industrial equipment monitoring order is selected as the initial solution, and the initial solution is set as the current solution to determine whether the current solution meets the constraint of the AI ​​robot monitoring time. If the constraint of the AI ​​robot monitoring time is met, determine the estimated value freq0 corresponding to the current solution, and set the initial temperature to the current temperature; if the constraint of the AI ​​robot monitoring time is not met, randomly select industrial equipment again to obtain the monitoring order of the industrial equipment; S52, set the counting variable num, and repeat steps S53 to S54 for num=1, 2, ..., N; N represents the number of times of the set inner loop; S53, by generating disturbances based on the current solution, changing the monitoring order of the industrial equipment, and determining whether the monitoring order of the industrial equipment after the disturbances are generated meets the constraint conditions of the AI ​​robot monitoring time; if the constraint conditions are met, the monitoring order of the industrial equipment after the disturbances is used as a new solution, and the sub-evaluation freq corresponding to the new solution is determined; if not, the disturbances are re-generated based on the current solution and the monitoring order of the industrial equipment is changed; S54, calculating the increment Δfreq of the sub-estimation value brought by the new solution, if the increment of the sub-estimation value is less than 0, accepting the new solution as the new current solution with probability 1, if the increment of the sub-estimation value is greater than or equal to 0, accepting the new solution as the new current solution with probability exp(-Δfreq / T), where T represents the current temperature; S55, after completing one iteration, select a cooling solution and lower the current temperature. If the current temperature is not less than the temperature threshold, enter step S52; if the current temperature is less than the threshold, determine the monitoring order of the industrial equipment according to the current solution.

6. A dynamic deployment method for AI robots according to claim 5, characterized in that: In step S51, the following steps are used to determine whether the current solution meets the constraints of the AI ​​robot monitoring time: S61, obtaining the time required for the AI ​​robot to move between two different industrial equipment and the time required for the AI ​​robot to complete one monitoring of each industrial equipment; S62, add the time required for the AI ​​robot to complete one monitoring of the industrial equipment included in the current solution to obtain the monitoring time of the current solution; add the time consumed by the AI ​​robot when moving according to the industrial equipment monitoring sequence corresponding to the current solution to obtain the movement time of the current solution; add the movement time and monitoring time of the current solution to obtain the time constraint of the current solution. If the time constraint of the current solution is not greater than the monitoring time Tset set by the AI ​​robot, the current solution meets the constraint of the AI ​​robot's monitoring time. Otherwise, the current solution does not meet the constraint of the AI ​​robot's monitoring time.

7. The dynamic deployment method for AI robots according to claim 5, characterized in that: Determining the industrial equipment causing the disturbance also includes the following steps: Let the monitoring order of industrial equipment corresponding to the current solution be Ax1, Ax2, ..., Axm, where Ax1, Ax2, ..., Axm represent industrial equipment and m is the number of industrial equipment monitored by the AI ​​robot; Ax1, Ax2, ..., Axm allow the existence of the same industrial equipment; Let the information increments of Ax1, Ax2, ..., Axm be Ipr respectively x1 , Ipr x2 ,…,Ipr xm , calculate the inverse of the information increment 1 / Ipr x1 , 1 / Ipr x2 ,…,1 / Ipr xm , the sum of the reciprocals of the information increment is taken as the length of the random interval, and m random subintervals are obtained according to the reciprocals of the information increment. The random subintervals are [0, Ipr x1 ]、[1 / Ipr x1 ,1 / Ipr x1 +1 / Ipr x2 ], …, [1 / Ipr x1 +…+1 / Ipr xm-1 ,1 / Ipr x1 +…+1 / Ipr xm ], the subintervals correspond one-to-one to the industrial equipment Ax1, Ax2, …, Axm; All random subintervals are concatenated into a random interval, and in the random interval [0,1 / Ipr x1 +…+1 / Ipr xm ]Generate a random number and find the industrial equipment corresponding to the random sub-interval where the random number is located. The industrial equipment will be replaced by a new industrial equipment; when the random number is at the end point of the random sub-interval, regenerate a random number to determine the replaced industrial equipment.

8. The dynamic deployment method for AI robots according to claim 7, characterized in that: Generate disturbances to industrial equipment by following these steps: S81, let the replaced industrial equipment be Axq, where q is a positive integer between the interval [1, m]; S82, under the monitoring sequence of the industrial equipment corresponding to the current solution, analyze the conditional information increment when the industrial equipment replaces the industrial equipment Axq at the position of the industrial equipment Axq, and let TIpr i represents the condition information increment of industrial equipment Ai, t i Indicates the time since the industrial equipment Ai was last monitored by the AI ​​robot. If t i If TIpr is less than the safety time of industrial equipment Ai, i =0; if t i Not less than the safety time of industrial equipment Ai, then TIpr i =Ipr i ×t i The safety time of the industrial equipment Ai refers to the time during which the parameter information of the industrial equipment Ai is consistent with the parameter information when the AI ​​robot monitors the industrial equipment Ai after the industrial equipment Ai is monitored by the AI ​​robot; S83, concatenate the condition information increments of the industrial equipment except Axq into an interval, and obtain M sub-intervals, where M+1 is the number of industrial equipment, and the sub-intervals are [0, TIpr1], [TIpr1, TIpr1+TIpr2], …, [TIpr1+…+TIpr M-1 ,TIpr1+…+TIpr M ], the sub-intervals correspond to the production information one by one; in the interval [0,TIpr1+…+TIpr M ]Generate a random number and determine the industrial equipment corresponding to the sub-interval where the random number is located. The industrial equipment will replace Axq to form a new solution; when the random number is at the endpoint of the sub-interval, a new random number is generated.

9. A dynamic deployment method for AI robots according to claim 8, characterized in that: In steps S51 and S53, the estimated values ​​of the monitoring sequence are determined by: The sum of the condition information increments ΣIpr of all industrial equipment in the monitoring sequence is obtained, and the condition information increment acquisition speed DV is calculated, DV=ΣIpr / Tset, and the reciprocal of the condition information increment acquisition speed DV 1 / DV is taken to obtain the estimated value of the monitoring sequence.

10. A dynamic deployment device for an AI robot, implementing a dynamic deployment method for an AI robot as claimed in any one of claims 1 to 9, characterized in that: include: AI robot module and data analysis module; the AI ​​robot module is used to control the AI ​​robot to move to the deployment position according to the monitoring sequence and perform a comprehensive detection of industrial equipment parameters; the data analysis module determines the deployment position of the AI ​​robot and the order of going to the deployment position when the AI ​​robot is monitoring.

Images