A big data computer room inspection method and system based on usage action frequency

Abstract

The present invention belongs to the technical field of computer room inspection, and specifically is a big data computer room inspection method and system based on the frequency of use of actions, including a preset inspection route; after the inspection robot arrives at the designated inspection position, it executes the detection action; the number of executions and the duration of each action of the inspection robot are recorded, and a data model is established; according to the data model, the action execution strategy and path planning of the inspection robot are optimized and an optimization mathematical model is established; the inspection is optimized according to the time scale of day / week / month, and a periodic mathematical model is established, and finally an optimized inspection plan is realized, so as to solve the problems of high energy consumption cost and short service life of the inspection robot in the prior art, and can minimize the inspection cost, reduce energy consumption and maximize the service life of the inspection robot according to the frequency of use of the detection module of the inspection robot.

Description

Technical Field

[0001] The present invention belongs to the technical field of computer room inspection, and specifically relates to a big data computer room inspection method and system based on usage action frequency. Background Art

[0002] With the explosive growth of data volume, data centers have become critical information infrastructure, and their stable operation is crucial to enterprise operations. Traditional manual inspection methods are time-consuming and labor-intensive, and it is difficult to ensure 24-hour uninterrupted and efficient monitoring. In order to improve inspection efficiency, reduce costs, and ensure the security and reliability of data centers, computer room inspection robots based on big data technology have been introduced. These robots can collect and analyze various environmental and equipment status data in the computer room in real time, improve response speed through automated inspections, achieve predictive maintenance, and reduce human errors. They use advanced sensors, artificial intelligence algorithms, and big data analysis to provide intelligent solutions for computer room operation and maintenance.

[0003] The inspection robot itself is equipped with multiple detection modules, including visual detection module, temperature detection module, humidity detection module, infrared thermal imaging module, etc. During a single inspection, the frequency of use of each detection module may vary depending on the specific needs of the computer room, the design of the inspection robot, and the inspection strategy. For example: temperature is one of the most critical parameters in computer room environmental monitoring, and the temperature detection module may need to be monitored frequently to ensure that the equipment does not overheat; the humidity detection module is similar to temperature monitoring, and humidity is also crucial to the safe operation of the equipment, so it may also be frequently detected; the infrared thermal imaging module is used to detect equipment overheating problems, and key equipment may be focused on during inspections.

[0004] However, actual usage frequency may be affected by the following factors:

[0005] Inspection strategy: The robot may adopt different inspection strategies. For example, it may only perform temperature checks in specific cabinets or areas.

[0006] Cabinet layout: If the distance between cabinets is very close, the robot may perform a one-time temperature check on adjacent cabinets;

[0007] Environmental changes: If the temperature distribution in the room is relatively uniform, the robot may reduce the detection frequency or perform temperature detection only when an abnormality is detected;

[0008] Sensor accuracy and range: If the temperature sensor has a wide detection range, the robot may not need to detect each cabinet individually;

[0009] Therefore, it is necessary to optimize the inspection strategy and layout of the inspection robot based on the usage frequency of each detection module in the inspection robot. It can minimize the inspection cost, reduce energy consumption, and maximize the service life of the inspection robot. To this end, we propose a big data computer room inspection method and system based on the frequency of use of actions. Summary of the invention

[0010] In order to solve the problems of high energy consumption cost and short service life of inspection robots in the prior art, the present invention provides a big data computer room inspection method and system based on usage frequency, which can minimize the inspection cost, reduce energy consumption and maximize the service life of the inspection robot according to the usage frequency of the detection module of the inspection robot.

[0011] To achieve the above purpose, the specific technical solution is as follows:

[0012] One of the present inventions is to provide a big data computer room inspection method based on the frequency of use of actions, including presetting an inspection route; after the inspection robot arrives at the designated inspection position, executing the detection action; recording the number of executions and duration of each action of the inspection robot, and establishing a data model; optimizing the action execution strategy and path planning of the inspection robot according to the data model and establishing an optimized mathematical model; optimizing inspections according to the time scale of day / week / month, and establishing a periodic mathematical model, and finally realizing the optimized inspection plan.

[0013] The second aspect of the present invention is to provide a big data computer room inspection system based on usage action frequency, comprising:

[0014] Data collection module: records the number of executions and duration of each action of the inspection robot and establishes a data model;

[0015] Optimization module of the inspection robot's motion execution strategy and path planning: define f(C i ,D i ) is the optimization objective function, which depends on action A i The number of executions C i and the average duration (total duration divided by number of executions);

[0016] Periodic optimization module: optimize inspections according to daily / weekly / monthly time scales, and adjust the execution frequency, sequence, and path planning of actions.

[0017] The third aspect of the present invention is to provide a big data computer room inspection robot based on the frequency of use of actions, and the device is communicatively connected to a big data computer room inspection system based on the frequency of use of actions.

[0018] A fourth aspect of the present invention is to provide a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in one of the purposes.

[0019] Compared with the prior art, the beneficial effects of the present invention are at least:

[0020] 1) By analyzing the frequency of each detection action performed by the robot, it helps determine which areas or equipment need closer monitoring, thereby achieving a more accurate and effective inspection plan. This not only improves the coverage and response speed of the inspection, but also reduces the risk of missing key issues;

[0021] 2) It can significantly improve inspection efficiency. By reducing unnecessary actions and adjusting the order of action execution, the robot can complete more tasks in a shorter time, reducing ineffective labor and energy consumption. This efficiency improvement directly translates into reduced operation and maintenance costs. At the same time, the robot's intelligent path planning reduces the moving distance, which helps to extend the robot's service life;

[0022] 3) Support sustainable development goals. By optimizing inspection routes and reducing energy consumption, data centers can reduce their impact on the environment and achieve green operations. At the same time, intelligent inspection processes reduce reliance on manual labor, reduce human resource costs, and improve the overall level of operation and maintenance management. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic diagram of the inspection optimization process of the inspection robot of the present invention. DETAILED DESCRIPTION

[0024] The present invention will be described in detail below in conjunction with the specific embodiments shown in the accompanying drawings. However, these embodiments do not limit the present invention, and any structural, methodological, or functional changes made by a person skilled in the art based on these embodiments are all within the scope of protection of the present invention.

[0025] like Figure 1 As shown, a big data computer room inspection method based on usage action frequency includes the following steps:

[0026] S1: Action definition: clarify all the detection actions that the inspection robot needs to perform, such as temperature detection, humidity detection, smoke detection, etc.;

[0027] S2: Data acquisition: Integrate sensors and data acquisition modules into the robot's control system to record the number of executions and duration of each action in real time;

[0028] S3: Frequency analysis: Regularly collect and analyze the robot's inspection data to determine which inspection actions are performed more frequently;

[0029] S4: Optimize inspection plan: dynamically adjust the robot's task scheduling according to the execution frequency and importance of the action, and optimize the inspection sequence and time;

[0030] S5: Optimize inspection paths: Use optimization algorithms to plan the robot’s inspection paths to ensure that high-frequency inspection actions are prioritized.

[0031] In the above step S1, the inspection actions performed by the inspection robot include: visual inspection, temperature inspection, humidity inspection, infrared thermal imaging, sound recognition, vibration detection, gas detection, smoke and fire detection, and electrical parameter detection.

[0032] In the above step S2, we record the number of executions and duration of each action of the inspection robot and establish a data model. We define the variables:

[0033] Let A i represents the i-th action, where i=1,2,...,N (N is the total number of actions that the inspection robot needs to perform),

[0034] Let T start,i Indicates action A i The time to start execution,

[0035] Let T end,i Indicates action A i The time to end execution,

[0036] Let D i Indicates action A i The duration can be expressed as:

[0037] D i =T end,i -T start,i

[0038] Let C i Indicates action A i The number of executions;

[0039] The specific recording process includes the following steps:

[0040] S201: For each action A i , initialize its execution count C i = 0 and duration D i =0;

[0041] S202: When action A i At the beginning, record the time T when execution starts start,i , and update the execution count: C i =C i +1;

[0042] S203: When action A i At the end, record the time T when the execution ends end,i , and calculate the duration :D i =D i + (T end,i -T start,i );

[0043] S204: C of each action i and D i stored in databases or log files;

[0044] S205: For the i-th execution of action A i , whose duration D i,k (The duration of the kth time) can be expressed as:

[0045] D i,k =T end,i,k -T start,i,k

[0046] Among them, T start,i,k and T end,i,k are the kth execution of action A i The start and end time of

[0047] S206: After each action is completed, the duration D i,k Accumulated to action A i Duration D i superior;

[0048] S207: Set a synchronization period P. At the end of this period, all C i and D i The current value of is synchronized to the central database;

[0049] The frequency analysis in step S3 above can be directly obtained from the data recorded in step S2 and recorded in a table. An example of the table is as follows:

[0050]

[0051] The above table records the data of each action when it is executed once. The data will change automatically when it is executed next time.

[0052] The optimized inspection plan and inspection path in step S4 and step S5 are implemented by using the collected data, including the following steps:

[0053] Step 1:

[0054] First, we perform optimization inspections on a daily / weekly / monthly basis and define the optimization parameters for each action:

[0055] C i :Action A i The number of executions per day / week / month,

[0056] D i :Action A i duration,

[0057] E i :Action A i Energy consumption,

[0058] Q i :Action A i The total importance score of

[0059] Step 2:

[0060] Create tables to collect and organize data:

[0061]

[0062] Step 3: Based on the table data, use linear programming to build an optimization model:

[0063] 1) Use an optimization model to minimize the total cost:

[0064] Cost(A i )=w1·C i +w2·

[0065] Cost(A i ) is action A i The cost function of , where w1 and w2 are weight coefficients;

[0066] Its constraints are:

[0067] T Max The maximum allowed time for a single inspection by the inspection robot.

[0068] N is the total number of actions that the inspection robot needs to perform,

[0069] The constraints can be expressed as:

[0070] ≦T Max

[0071] The optimization mathematical model can be expressed as:

[0072] Minw1,w2

[0073] st ≦T Max

[0074] In table form, it is:

[0075]

[0076] Based on the table, identify actions with high execution times or long average duration, and for actions with high costs, consider technical improvements or process optimization.

[0077] 2) The table in step 2 is periodic data. The optimization model is used with periodic data to minimize the total cost (time, energy consumption) and maximize the detection quality:

[0078] Min( )

[0079] Max( )

[0080] Where w1 and w2 are weight coefficients

[0081] The total inspection time should not exceed the maximum allowed time per day / week / month, i.e.

[0082] ≦T Max

[0083] According to the results of the optimization algorithm, adjust the execution frequency, sequence and path planning of the actions

[0084] Assuming that we find that the energy consumption E1 of action A1 is high through the optimization model, we can consider:

[0085] ① Reduce the number of executions of action A1;

[0086] ② Find alternative low-energy actions;

[0087] ③Perform action A1 during off-peak hours;

[0088] 3) Establish an optimization model to minimize total energy consumption or maximize detection coverage:

[0089] Min=

[0090] Max=

[0091] Its constraints are those that define inspection time and energy consumption:

[0092] ≦T Max ≦E Max

[0093] ≧Q Min

[0094] That is, the total time does not exceed the maximum allowed time, ensuring that the total importance score does not fall below the minimum threshold;

[0095] According to the optimization results, a detailed inspection plan is formulated, including the execution sequence and time schedule of the actions;

[0096] The example table of the inspection action optimization plan is as follows:

[0097]

[0098] By optimizing models and examples, it is possible to ensure that energy consumption is reduced and the service life of the robot is increased while ensuring inspection quality and efficiency.

[0099] Step 4: The weight coefficients w1 and w2 in step 3 are calculated using the entropy weight method, which specifically includes the following steps:

[0100] 1) Assume we have N actions, and each action is executed C times i , the duration is D i (i=1,2,3,...,N), we want to determine the weight coefficient wi for each action;

[0101] 2) First, we need to standardize the number of executions and duration to eliminate the impact of different dimensions. We use the range standardization method:

[0102]

[0103]

[0104] in, and are the normalized values ​​of execution times and duration, respectively. is the maximum and minimum number of times all actions are executed. are the maximum and minimum durations of all actions;

[0105] We calculate the entropy value of each indicator, which reflects the degree of discreteness of the indicator:

[0106]

[0107] in, = or = , depending on whether we consider the number of executions or the duration;

[0108] The coefficient of variation gi represents the effective information of indicator i and is calculated as follows:

[0109]

[0110] Determine the weight coefficient for each action :

[0111]

[0112] Based on the above steps, we use the linear programming model as the algorithm for optimizing the model, which includes the following steps:

[0113] According to our goal, we set the objective function as:

[0114] +

[0115] Where Z is the total cost, and is the weight coefficient used to balance time and energy consumption, T is the total time cost, and E is the total energy cost;

[0116] The constraints are:

[0117]

[0118]

[0119]

[0120] in, is the average execution time of action i,

[0121] is the energy consumption of action i,

[0122] is the daily execution frequency of action i,

[0123] is the maximum allowed time for daily inspections,

[0124] It is the maximum allowable energy consumption for daily inspection.

[0125] According to the above steps, we give an example:

[0126] Suppose we have three actions, and their execution time and energy consumption are as follows:

[0127]

[0128] We hope to minimize the total cost while ensuring that the maximum daily time does not exceed 30 minutes and the maximum energy consumption does not exceed 15 units. Using linear programming, we can establish the following model:

[0129] +

[0130] Constraints:

[0131] 5 +3 +4 30

[0132] 2 +1 +3 15

[0133] , , 0

[0134] By solving the above linear programming problem, we can get the optimal execution frequency of each action , , , thereby optimizing the inspection strategy of the inspection robot.

[0135] Embodiment 2:

[0136] A big data computer room inspection system based on usage action frequency, comprising:

[0137] Data collection module: records the number of executions and duration of each action of the inspection robot and establishes a data model;

[0138] Optimization module of the inspection robot's motion execution strategy and path planning: define f(C i ,D i ) is the optimization objective function, which depends on action A i The number of executions C i and the average duration (total duration divided by number of executions);

[0139] Periodic optimization module: optimize inspections according to daily / weekly / monthly time scales, and adjust the execution frequency, sequence, and path planning of actions.

[0140] Embodiment three:

[0141] A big data computer room inspection robot based on usage action frequency is communicatively connected with a big data computer room inspection system based on usage action frequency.

[0142] Embodiment 4:

[0143] A computer readable medium stores a computer program, which, when executed by a processor, implements the method as described in one of the purposes.

[0144] It should be understood that although the present specification is described according to embodiments, not every embodiment contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

[0145] The series of detailed descriptions listed above are only specific descriptions of feasible implementation methods of the present invention. They are not intended to limit the scope of protection of the present invention. Any equivalent implementation methods or changes that do not deviate from the technical spirit of the present invention should be included in the scope of protection of the present invention.

Claims

1. A big data computer room inspection method based on usage action frequency, characterized in that: include: Preset inspection routes; After the inspection robot reaches the designated inspection position, it performs the inspection action; To record the number of times and duration of each action of the inspection robot and build a data model, we define the variables: Let A i represents the i-th action, where i=1,2,...,N; N is the total number of actions that the inspection robot needs to perform; Let T start,i Indicates action A i The time to start execution, Let T end,i Indicates action A i The time to end execution, Let D i Indicates action A i The duration of is expressed as: D i =T end,i -T start,i Let C i Indicates action A i The number of executions; According to the data model, the action execution strategy and path planning of the inspection robot are optimized and the optimization mathematical model is established. i ,D i ) is the optimization objective function, which depends on action A i The number of executions C i and the average duration ; We optimize inspections on a daily / weekly / monthly basis and establish a periodic mathematical model. We define the optimization parameters for each action: C i :Action A i The number of executions is the number of executions per day, week, or month. D i :Action A i duration, E i :Action A i Energy consumption, Q i :Action A i The total importance score of Establish a mathematical model function that minimizes total energy consumption and maximizes coverage: Min= Max= The inspection actions performed by the inspection robot include: visual inspection, temperature inspection, humidity inspection, infrared thermal imaging, sound recognition, vibration detection, gas detection, smoke and fire detection and electrical parameter detection; Record the execution times and duration of each action of the inspection robot, including the following steps: S1: For each action A i , initialize its execution count C i = 0 and duration D i =0; S2: When action A i At the beginning, record the time T when execution starts start,i , and update the execution count: C i =C i +1; S3: When action A i At the end, record the time T when the execution ends end,i , and calculate the duration :D i =D i + (T end,i -T start,i ); S4: C of each action i and D i stored in databases or log files; S5: For the i-th execution of action A i , whose duration D i,k It is expressed as: D i,k =T end,i,k -T start,i,k S6: After each action is completed, the duration D i,k Accumulated to action A i Duration D i superior; S7: Set a synchronization period P. At the end of this period, all C i and D i The current value of is synchronized to the central database; The f(C i ,D i ) is the optimization objective function, which depends on action A i The number of executions C i and the average duration , the optimization mathematical model also includes a cost function: Cost(A i )=w1·C i +w2· Cost(A i ) is action A i The cost function of , where w1 and w2 are weight coefficients; The optimization mathematical model also includes constraints: T Max The maximum allowed time for a single inspection by the inspection robot. N is the total number of actions that the inspection robot needs to perform, The constraints are expressed as: ≦T Max The optimization mathematical model is expressed as: Minw1,w2 s.t. ≦T Max The constraints of the mathematical model function for minimizing total energy consumption and maximizing coverage are the constraints that define inspection time and energy consumption: ≦T Max ≦E Max Q i ≧Q Min That is, the total time does not exceed the maximum allowed time, ensuring that the total importance score does not fall below the minimum threshold; Among them, C i It is action A i The number of executions, D i It is action A i The duration of E i It is action A i Energy consumption, T Max is the maximum allowed time for daily inspection, Q Min is the minimum threshold of the total importance score; The weight coefficients w1, w2 are calculated using the entropy weight method.

2. A big data computer room inspection system based on usage action frequency, having an application in a big data computer room inspection method based on usage action frequency as claimed in claim 1, characterized in that: include: Data collection module: records the number of executions and duration of each action of the inspection robot and establishes a data model; Optimization module of the inspection robot's motion execution strategy and path planning: define f(C i ,D i ) is the optimization objective function, which depends on action A i The number of executions C i and the average duration ; Periodic optimization module: optimize inspections according to daily / weekly / monthly time scales, and adjust the execution frequency, sequence, and path planning of actions.

3. A big data room inspection robot based on usage action frequency, characterized in that: The big data computer room inspection robot based on usage action frequency is communicatively connected to the big data computer room inspection system based on usage action frequency described in claim 2.

4. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to claim 1 is implemented.

Images