Intelligent management method and system of plan review robot based on multi-source data analysis

Abstract

The present invention relates to the field of data processing technology, and specifically discloses an intelligent management method and system for a plan review robot based on multi-source data analysis, wherein the review status value of the plan review robot is obtained according to the review status data of the plan review robot in the review process; the review status value of the plan review robot and the review status comparison value of the plan review robot are analyzed and processed to obtain a review status signal of the plan review robot within a monitoring period; based on the review status signal, the hardware status of the plan review robot within the monitoring period is monitored and identified to complete visual identification of hardware control. The present invention completes the division of the review performance of the plan review robot within the monitoring period by identifying the review status of the plan review robot within the monitoring period, which facilitates subsequent adjustment and processing of the plan review robot according to the multi-level status division of the plan review robot.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a plan review robot intelligent management method and system based on multi-source data analysis. Background Art

[0002] The plan review robot is an intelligent system that integrates multi-source data analysis, machine learning, natural language processing and other technologies, and can assist or replace manual work in completing complex plan review tasks. The system provides comprehensive and accurate data support to decision makers by collecting, integrating and analyzing information from different data sources, while using intelligent algorithms to automatically perform review tasks and improve review efficiency and accuracy.

[0003] At present, in the working process of the plan review robot, more emphasis is placed on improving the efficiency and accuracy of plan review by integrating and analyzing multi-source data. There is a lack of performance monitoring of the plan review robot during long-term processing, and the status identification of the plan review robot cannot be realized in a timely and effective manner, which has certain risks and limitations.

[0004] Based on this, an intelligent management method and system of plan review robots based on multi-source data analysis are proposed to optimize the entire review process. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent management method and system for a plan review robot based on multi-source data analysis. When the review status of the plan review robot is not good or unstable, the risk screening value (performance) of the electronic components of the plan review robot in the process of each document review within the monitoring period is obtained, and the risk screening value of the electronic components is extended to the entire monitoring period, thereby improving the accuracy of the risk screening of electronic components, facilitating better hardware control of the plan review robot with high accuracy, so as to solve the problems in the above background.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The intelligent management method of the plan review robot based on multi-source data analysis includes the following steps:

[0008] Obtain the review status data of the plan review robot during the review process;

[0009] Among them, the review status data includes the review external explicit analysis value, review internal factor analysis value and review external disturbance analysis value of the review document by the planned review robot;

[0010] Based on the review status data of the plan review robot, obtaining the review status value of the plan review robot;

[0011] During the monitoring period, the review status value of the plan review robot and the review status comparison value of the plan review robot are analyzed and processed to obtain the review status signal of the plan review machine during the monitoring period;

[0012] The review status signal includes a primary status signal, a secondary status signal, a tertiary status signal and a quaternary status signal of the planned review robot;

[0013] Based on the review status signal, the hardware status of the planned review robot during the monitoring cycle is monitored and identified to complete the visual identification of hardware management and control.

[0014] As a further solution of the present invention: the process of obtaining the review status value of the planned review robot is as follows:

[0015] The censored external analysis value is recorded as Cw;

[0016] The internal factor analysis value of the review is recorded as Cn;

[0017] The external disturbance value is recorded as Cr;

[0018] By formula The review status value Ci of the plan review robot is calculated, where Si is the performance status value of the plan review robot.

[0019] As a further solution of the present invention: the process of obtaining the external analysis value is as follows:

[0020] Get the time from when the planned review robot starts reviewing the review document to when the review report is generated, and record it as the total review time;

[0021] Calculate the ratio of the total review time to the size of the review file to obtain the review external analysis value of the planned review robot;

[0022] The process of obtaining the internal factor analysis value is as follows:

[0023] Obtain the parsing speed and compression speed of the review file by the plan review robot, sum and average the parsing speed and compression speed of the review file by the plan review robot, and obtain the internal factor analysis value of the review file;

[0024] The process of obtaining the external disturbance value is as follows:

[0025] The total review time is divided into several time sub-units, and the network delay time of the planned review robot when reviewing the review file is obtained through the Ping test in each time sub-unit;

[0026] Sum up the network delay times of all time subunits to obtain the total review delay time;

[0027] The ratio of the total review delay time to the review file size is calculated to obtain the review out-of-review disturbance value of the review file.

[0028] As a further solution of the present invention: the process of obtaining the effectiveness status value of the planning review robot is as follows:

[0029] Get the capacity effectiveness ratio of the plan review robot, denoted as Sr;

[0030] Get the maximum throughput of the plan review robot, denoted as St;

[0031] Get the historical packet loss rate of the plan review robot, denoted as Sd;

[0032] By formula The efficiency state value Si of the plan review robot is calculated, where a1 and a2 are preset proportional coefficients.

[0033] As a further solution of the present invention: obtaining the review state comparison value of the planned review robot before the monitoring cycle, recorded as C0;

[0034] Obtain the review status value Ci of the plan review robot during the monitoring period, where i is the number of reviews of the plan review robot;

[0035] Traverse the review status values ​​obtained by the review of all review files by the planned review robot within the monitoring period to obtain the maximum review status value Cmax and the minimum review status value Cmin;

[0036] If the maximum review state value Cmax ≤ the review state comparison value C0 within the monitoring period, a level 4 state signal of the plan review robot is generated;

[0037] If the minimum review state value Cmin within the monitoring period is greater than the review state comparison value C0, a primary state signal of the plan review robot is generated;

[0038] If the maximum review state value Cmax within the monitoring period is greater than the review state comparison value C0 and greater than the minimum review state value Cmin within the monitoring period, a first-level verification state signal of the plan review robot is generated.

[0039] As a further solution of the present invention: based on the primary verification state signal of the plan review robot, the review state variance value of the plan review robot within the monitoring period is obtained, which is recorded as Cf;

[0040] When the review state variance value Cf ≥ the review state variance threshold Cy, a secondary state signal of the plan review robot is generated;

[0041] When the review state variance value Cf is less than the review state variance threshold Cy, a secondary verification state signal of the plan review robot is generated.

[0042] As a further solution of the present invention: based on the secondary validation status signal of the plan review robot, the review status value of the plan review robot within the monitoring period is identified;

[0043] If the review state value Ci of the plan review robot decreases linearly and continuously during the monitoring period, a third-level state signal of the plan review robot is generated;

[0044] If the review state value Ci of the plan review robot increases linearly and continuously during the monitoring period, a primary state signal of the plan review robot is generated.

[0045] As a further solution of the present invention: within the monitoring cycle, the hardware status monitoring and identification includes:

[0046] Obtain the electronic components that the planned review robot needs to monitor when reviewing documents;

[0047] Record the items that need to be monitored for electronic components as monitoring items;

[0048] The monitoring items in the electronic components whose monitoring values ​​do not meet the monitoring value requirements of the monitoring items are recorded as risk items;

[0049] Process the risk items to obtain the total risk value and risk item ratio of the electronic components;

[0050] The total risk value of the electronic components is multiplied by the risk item ratio to obtain the risk screening value of the electronic components;

[0051] The risk screening values ​​obtained each time the documents of the same electronic component are reviewed are summed and averaged to obtain the risk screening analysis value of the electronic component during the monitoring period.

[0052] As a further solution of the present invention: the risk screening analysis values ​​of all electronic components that need to be monitored in the plan review robot are sorted, and the risk screening analysis values ​​are sorted in descending order to obtain the risk levels of the corresponding electronic components and conduct screening and processing.

[0053] As a further solution of the present invention: a plan review robot intelligent management system based on multi-source data analysis includes:

[0054] A data analysis module, which is used to obtain the review status data of the plan review robot during the review process and transmit the review status data to the cloud management and control platform;

[0055] A status verification module, wherein the status verification module receives the review status data transmitted by the cloud management and control platform, processes the review status data, obtains the review status value of the planned review robot, and sends the review status value to the cloud management and control platform;

[0056] A multimodal analysis module, wherein the multimodal analysis module receives the review status value transmitted by the cloud management and control platform. During the monitoring period, the multimodal analysis module analyzes and processes the review status value of the plan review robot and the review status comparison value of the plan review robot to obtain the review status signal of the plan review machine during the monitoring period, and transmits the review status signal to the cloud management and control platform;

[0057] The review status signal includes a primary status signal, a secondary status signal, a tertiary status signal and a quaternary status signal of the planned review robot;

[0058] The performance control module receives the review status signal transmitted by the cloud control platform. The performance control module monitors and identifies the hardware status of the planned review robot during the monitoring period, and completes the visual identification of hardware control.

[0059] Beneficial effects of the present invention:

[0060] The present invention collects the review status data of the plan review robot in the process of reviewing documents, that is, by processing the total review time of the plan review robot, the parsing speed of the review document, the compression speed of the review document and the network delay time when the review document is reviewed, the review external explicit analysis value, the review internal factor analysis value and the review external disturbance analysis value of the review document are obtained, and the review status value of the review robot is planned according to the review external explicit analysis value, the review internal factor analysis value and the review external disturbance analysis value of the review document, so as to realize the recognition of the plan review robot status from the dimensions of the external environment and internal factors of the plan review robot in processing the review document, with high accuracy;

[0061] The present invention compares the review status values ​​of the plan review robot multiple times within the monitoring period with the review status comparison value of the plan review robot, thereby realizing the identification of the review status of the plan review robot within the monitoring period based on the review status comparison value, completing the division of the review performance of the plan review robot within the monitoring period, and facilitating the subsequent adjustment and processing of the plan review robot according to the multi-level status division of the plan review robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The present invention will be further described below in conjunction with the accompanying drawings.

[0063] Figure 1 is a flow chart of a plan review robot intelligent management method based on multi-source data analysis according to an embodiment of the present invention;

[0064] Figure 2 It is a state recognition flow chart of a plan review robot intelligent management method based on multi-source data analysis in an embodiment of the present invention;

[0065] Figure 3 It is a flowchart of the program of the plan review robot intelligent management system based on multi-source data analysis in the embodiment of the present invention. DETAILED DESCRIPTION

[0066] 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.

[0067] Example 1, please refer to Figure 1 As shown, the present invention is a plan review robot intelligent management method based on multi-source data analysis, comprising the following steps:

[0068] Obtain the review status data of the plan review robot during the review process;

[0069] Based on the review status data of the plan review robot, obtaining the review status value of the plan review robot;

[0070] During the monitoring period, the review status value of the plan review robot and the review status comparison value of the plan review robot are analyzed and processed to obtain the review status signal of the plan review machine during the monitoring period;

[0071] The review status signal includes a primary status signal, a secondary status signal, a tertiary status signal and a quaternary status signal of the planned review robot;

[0072] Based on the review status signal, the hardware status of the planned review robot during the monitoring cycle is monitored and identified to complete the visual identification of hardware management and control.

[0073] The review status data includes the external explicit analysis value, internal factor analysis value and external disturbance analysis value of the review document by the planned review robot;

[0074] The censored external analysis value is recorded as Cw;

[0075] The internal factor analysis value of the review is recorded as Cn;

[0076] The external disturbance value is recorded as Cr;

[0077] By formula The review status value Ci of the plan review robot is calculated, where Si is the performance status value of the plan review robot.

[0078] The process of obtaining the external analysis value is as follows:

[0079] Get the time from when the planned review robot starts reviewing the review document to when the review report is generated, and record it as the total review time;

[0080] The total review time is calculated by ratio with the size of the reviewed file to obtain the review explicit analysis value of the planned review robot.

[0081] The process of obtaining the internal factor analysis value is as follows:

[0082] Obtain the parsing speed and compression speed of the review file by the plan review robot, sum and average the parsing speed and compression speed of the review file by the plan review robot, and obtain the internal factor analysis value of the review file;

[0083] The parsing speed of the review file is the ratio of the parsing time of the review robot during the review process of the review file to the size of the review file;

[0084] Among them, the compression speed of the review file is the ratio of the compression time of the review robot during the review process of the review file to the size of the review file.

[0085] Among them, the process of obtaining the external disturbance value is as follows:

[0086] The total review time is divided into several time sub-units, and the network delay time of the planned review robot when reviewing the review file is obtained through the Ping test in each time sub-unit;

[0087] Sum up the network delay times of all time subunits to obtain the total review delay time;

[0088] The ratio of the total review delay time to the review file size is calculated to obtain the review out-of-review disturbance value of the review file.

[0089] Through the above formula for obtaining the review status value of the plan review robot, it can be seen that the smaller the review external analysis value is, that is, the shorter the time taken by the plan review robot to review the document, the faster the review speed of the plan review robot is, the larger the review status value of the plan review robot is, and the higher the review efficiency of the plan review robot is;

[0090] The smaller the internal factor analysis value of the review, that is, the faster the parsing speed (ratio of parsing time to the size of the review file) and compression speed (ratio of compression time to the size of the review file) of the plan review robot when reviewing files, the larger the review status value of the plan review robot is, and the smoother the plan review robot is in the review process;

[0091] The smaller the external disturbance value of the review is, that is, the smaller the delay time of the plan review robot when reviewing documents, the smoother the plan review robot is when reviewing documents, and the larger the review status value of the plan review robot is.

[0092] The process of obtaining the effectiveness status value of the plan review robot is as follows:

[0093] Get the capacity effectiveness ratio of the plan review robot, denoted as Sr;

[0094] Get the maximum throughput of the plan review robot, denoted as St;

[0095] Get the historical packet loss rate of the plan review robot, denoted as Sd;

[0096] By formula The efficiency state value Si of the plan review robot is calculated, where a1 and a2 are preset proportional coefficients, and a1 and a2 are both greater than zero;

[0097] Among them, the efficiency status value of the plan review robot represents the running status of the plan review robot. The above formulas are obtained by collecting a large amount of data for software simulation and a formula close to the real value is selected. The proportional coefficient in the formula is set by technical personnel in this field according to actual conditions. The technical personnel in this field collect the capacity effective ratio of multiple groups of plan review robots, the maximum throughput of the plan review robots and the historical packet loss rate of the plan review robots, and process the capacity effective ratio of multiple groups of plan review robots, the maximum throughput of the plan review robots and the historical packet loss rate of the plan review robots to obtain the efficiency status value of the plan review robot respectively, and the calculated proportional coefficient is screened and averaged to obtain a1 and a2;

[0098] The size of the proportionality coefficient is to quantify each parameter to obtain a specific value for subsequent comparison. The size of the proportionality coefficient depends on the capacity effectiveness ratio of the plan review robot, the maximum throughput of the plan review robot, the historical packet loss rate of the plan review robot, and the technical personnel in this field preliminarily set the corresponding performance status value of the plan review robot for each set of data; as long as it does not affect the proportional relationship between the parameter and the quantified value, such as the capacity effectiveness ratio of the electric plan review robot is proportional to the performance status value of the plan review robot.

[0099] The effective capacity ratio of the planned review robot is the ratio of the remaining memory capacity of the planned review robot each time it reviews a document to the maximum capacity of the planned review robot;

[0100] Among them, the maximum throughput of the plan review robot can be tested by using the sysbench tool, LoadRunner tool or OracleSQLDeveloper tool;

[0101] The process of obtaining the historical packet loss rate of the plan review robot is as follows: obtaining the packet loss rate of each file reviewed by the plan review robot before the current time, and summing up the packet loss rates of all the files reviewed, that is, obtaining the historical packet loss rate of the plan review robot;

[0102] The process of obtaining the packet loss rate of the review file is as follows: before reviewing the file, obtaining the number of data packets received by the planned review robot; after reviewing the file, obtaining the number of data packets sent by the planned review robot;

[0103] The difference between the number of data packets received by the plan review robot and the number of data packets sent by the plan review robot is calculated to obtain a data packet difference;

[0104] The packet loss rate of the reviewed file is obtained by calculating the ratio of the packet difference to the number of packets received by the planned review robot.

[0105] Embodiment 2, in the monitoring period, obtain the review state value Ci of the plan review robot, i is the review times of the plan review robot, i=1, ..., n;

[0106] Obtain the review status value of the planned review robot before the monitoring cycle, and record it as the review status comparison value of the planned review robot, recorded as C0;

[0107] It should be noted that the review status comparison value of the plan review robot is the average of one or more review status values ​​before the monitoring period, and is assessed by the staff based on experience.

[0108] Traverse the review status values ​​obtained by the review of all review files by the planned review robot within the monitoring period to obtain the maximum review status value Cmax and the minimum review status value Cmin;

[0109] See also Figure 2 , if the maximum review state value Cmax within the monitoring period ≤ the review state comparison value C0, it means that the review state of the plan review robot is poor within the monitoring period, and a four-level state signal of the plan review robot is generated;

[0110] If the minimum review state value Cmin in the monitoring period is greater than the review state comparison value C0, it means that the review state of the plan review robot is good in the monitoring period, and a first-level state signal of the plan review robot is generated;

[0111] If the maximum review state value Cmax within the monitoring period is greater than the review state comparison value C0 and greater than the minimum review state value Cmin within the monitoring period, it means that the review state of the plan review robot has fluctuated within the monitoring period, and a first-level validation state signal of the plan review robot is generated;

[0112] Based on the primary verification status signal of the plan review robot, all the review status values ​​of the plan review robot within the monitoring period are processed to obtain the review status variance value of the plan review robot within the monitoring period, which is recorded as Cf;

[0113] Compare the review state variance value Cf with the review state variance threshold Cy;

[0114] When the review state variance value Cf ≥ the review state variance threshold Cy, it means that the review state of the plan review robot fluctuates greatly during the monitoring period, and a secondary state signal of the plan review robot is generated;

[0115] When the review state variance value Cf is less than the review state variance threshold Cy, it means that the review state fluctuation of the plan review robot is small during the monitoring period, and a secondary validation state signal of the plan review robot is generated;

[0116] Based on the secondary verification status signal of the plan review robot, the review status value of the plan review robot within the monitoring period is identified;

[0117] If the review state value Ci of the plan review robot decreases linearly and continuously during the monitoring period, it means that the review state of the plan review robot gradually deteriorates during the monitoring period, and a third-level state signal of the plan review robot is generated;

[0118] If the review state value Ci of the plan review robot increases linearly and continuously during the monitoring period, it means that the review state of the plan review robot is gradually improving during the monitoring period, and a first-level state signal of the plan review robot is generated;

[0119] If the review status value Ci of the planned review robot during the monitoring period is not linearly related, it means that the review status of the planned review robot during the monitoring period is in dynamic balance (the fluctuation of the review status value is formed according to the content of the review document).

[0120] Among them, the monitoring period is set by the staff based on experience, including but not limited to 3 days, 30 days and 90 days.

[0121] It should be noted that the first-level status signal, the second-level status signal, the third-level status signal and the fourth-level status signal are the divisions of the review status of the plan review robot within the monitoring period. The multi-level status division of the plan review robot is convenient for subsequent adjustment and processing of the plan review robot. Among them, the review status corresponding to the first-level status signal of the plan review robot is higher than the review status corresponding to the second-level status signal, the review status corresponding to the second-level status signal of the plan review robot is higher than the review status corresponding to the third-level status signal, and the review status corresponding to the third-level status signal of the plan review robot is higher than the review status corresponding to the fourth-level status signal.

[0122] Embodiment 3, based on the multi-level status signal of the plan review robot, the hardware status of the plan review robot is monitored and controlled, specifically:

[0123] It should be noted that: when the review status signal of the plan review robot is not a primary status signal, it means that the review status of the plan review robot is not good or unstable, and it is necessary to analyze and deal with the interference factors of the bad or unstable review status of the plan review robot;

[0124] This embodiment takes the status signal of the plan review robot being in the fourth level status signal as an example:

[0125] Obtain the electronic components that the planned review robot needs to monitor when reviewing documents;

[0126] Record the items that need to be monitored for electronic components as monitoring items;

[0127] During the monitoring cycle, the monitoring value of each monitoring item in the electronic components is collected during each review of the documents, and the monitoring value of the corresponding monitoring item is compared with the monitoring value requirement of the monitoring item;

[0128] The monitoring items in the electronic components whose monitoring values ​​do not meet the monitoring value requirements of the monitoring items are recorded as risk items;

[0129] The monitoring items in the electronic components whose monitoring values ​​meet the monitoring value requirements of the monitoring items are recorded as stable items;

[0130] Obtain the number of risk items in the electronic components, calculate the ratio of the number of risk items to the number of monitoring items, and obtain the risk item ratio of the electronic components;

[0131] Obtain the deviation value of the real-time data of the risk item compared with the required monitoring value of the monitoring item, which is recorded as the deviation difference of the risk item;

[0132] Obtain the preset risk coefficient of the corresponding risk item, multiply the deviation difference of the risk item by the preset risk coefficient to obtain the risk analysis value;

[0133] Among them, the values ​​of the preset risk coefficients are all greater than zero. The larger the value of the preset risk coefficient is, the greater the impact of the deviation of the corresponding risk item on the instability of the running state of the plan review robot is;

[0134] Obtain the risk analysis value of each risk item in the electronic components, and sum them up to obtain the total risk value of the electronic components;

[0135] The total risk value of the electronic components is multiplied by the risk item ratio to obtain the risk screening value of the electronic components;

[0136] During the monitoring period, the risk screening values ​​obtained each time the documents of the same electronic component are reviewed are summed and averaged to obtain the risk screening analysis value of the electronic component during the monitoring period;

[0137] The risk investigation and analysis values ​​of all electronic components that need to be monitored in the plan review robot are sorted, and the risk investigation and analysis values ​​are sorted in descending order to obtain the risk levels of the corresponding electronic components. The electronic components are investigated and processed according to the risk investigation and analysis values ​​in descending order to complete the hardware control of the plan review robot.

[0138] Among them, the planned review robot includes but is not limited to sensors (visual sensors, sound sensors and distance sensors), actuators (servos, motors), computing units (multi-core processors), power systems and expandable interfaces;

[0139] The monitoring items of electronic components include but are not limited to current value, voltage value, temperature value, vibration frequency, etc. The monitoring content of the monitoring items is determined according to the corresponding usage scenarios of the electronic components. The content of the monitoring items is generated by the corresponding components when they are working. For example, the current value and temperature value are the actual current value and temperature value when the monitoring items are working.

[0140] Example 4, please refer to Figure 3 As shown, the present invention is a plan review robot intelligent management system based on multi-source data analysis, including a data analysis module, a status verification module, a multimodal analysis module, a performance control module and a cloud control platform;

[0141] The data analysis module is used to obtain the review status data of the plan review robot during the review process and transmit the review status data to the cloud management and control platform;

[0142] The status verification module receives the review status data transmitted by the cloud management and control platform, processes the review status data, obtains the review status value of the planned review robot, and sends the review status value to the cloud management and control platform;

[0143] The multimodal analysis module receives the review status value transmitted by the cloud control platform. During the monitoring period, the multimodal analysis module analyzes and processes the review status value of the plan review robot and the review status comparison value of the plan review robot to obtain the review status signal of the plan review machine during the monitoring period, and transmits the review status signal to the cloud control platform;

[0144] The review status signal includes a primary status signal, a secondary status signal, a tertiary status signal and a quaternary status signal of the planned review robot;

[0145] The performance control module receives the review status signal transmitted by the cloud control platform. The performance control module monitors and identifies the hardware status of the planned review robot during the monitoring period, and completes the visual identification of hardware control.

[0146] One of the core points of the present invention is to collect the review status data of the planned review robot in the process of reviewing documents, that is, by processing the total time of the planned review robot reviewing documents, the parsing speed of the review documents, the compression speed of the review documents and the network delay time when the review documents are reviewed, the review external explicit analysis value, the review internal factor analysis value and the review external disturbance analysis value of the review document are obtained, and the review status value of the review robot is planned according to the review external explicit analysis value, the review internal factor analysis value and the review external disturbance analysis value of the review document, so as to realize the recognition of the planned review robot status from the dimensions of the external environment (network delay) and internal factors (parsing compression) of the planned review robot in the processing of the review documents, with high accuracy;

[0147] One of the core points of the present invention is to compare the review status values ​​of the plan review robot multiple times in the monitoring cycle with the review status comparison value of the plan review robot, so as to realize the recognition of the review status of the plan review robot in the monitoring cycle based on the review status comparison value, complete the division of the review performance of the plan review robot in the monitoring cycle, and facilitate the subsequent adjustment and processing of the plan review robot according to the multi-level status division of the plan review robot;

[0148] One of the core points of the present invention is that the review status signal of the plan review robot is not a primary status signal. When the review status of the plan review robot is not good or unstable, the risk screening value (performance) of the electronic components of the plan review robot in the process of each document review within the monitoring period is obtained, and the risk screening value of the electronic components is extended to the entire monitoring period, thereby improving the accuracy of the risk screening of electronic components, facilitating better hardware control of the plan review robot with high accuracy.

[0149] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A plan review robot intelligent management method based on multi-source data analysis, characterized in that: The following steps are involved: Obtain the review status data of the plan review robot during the review process; Among them, the review status data includes the review external explicit analysis value, review internal factor analysis value and review external disturbance analysis value of the review document by the planned review robot; Based on the review status data of the plan review robot, obtaining the review status value of the plan review robot; During the monitoring period, the review status value of the plan review robot and the review status comparison value of the plan review robot are analyzed and processed to obtain the review status signal of the plan review machine during the monitoring period; The review status signal includes a primary status signal, a secondary status signal, a tertiary status signal and a quaternary status signal of the planned review robot; Based on the review status signal, the hardware status of the planned review robot within the monitoring cycle is monitored and identified, completing the visual identification of hardware control; The process of obtaining the review status value of the planned review robot is as follows: The censored external analysis value is recorded as Cw; The internal factor analysis value of the review is recorded as Cn; The external disturbance value is recorded as Cr; By formula The review state value Ci of the plan review robot is calculated, where Si is the efficiency state value of the plan review robot; The process of obtaining the external analysis value is as follows: Get the time from when the planned review robot starts reviewing the review document to when the review report is generated, and record it as the total review time; Calculate the ratio of the total review time to the size of the review file to obtain the review external analysis value of the planned review robot; The process of obtaining the internal factor analysis value is as follows: Obtain the parsing speed and compression speed of the review file by the plan review robot, sum and average the parsing speed and compression speed of the review file by the plan review robot, and obtain the internal factor analysis value of the review file; The process of obtaining the external disturbance value is as follows: The total review time is divided into several time sub-units, and the network delay time of the planned review robot when reviewing the review file is obtained through the Ping test in each time sub-unit; Sum up the network delay times of all time subunits to obtain the total review delay time; The total review delay time is calculated by the ratio of the review file size to obtain the review out-of-review disturbance value of the review file; The process of obtaining the effectiveness status value of the plan review robot is as follows: Get the capacity effectiveness ratio of the plan review robot, denoted as Sr; Get the maximum throughput of the plan review robot, denoted as St; Get the historical packet loss rate of the plan review robot, denoted as Sd; By formula The efficiency state value Si of the plan review robot is calculated, where a1 and a2 are preset proportional coefficients.

2. The intelligent management method of plan review robot based on multi-source data analysis according to claim 1 is characterized in that: Obtain the review status comparison value of the planned review robot before the monitoring cycle, recorded as C0; Obtain the review status value Ci of the plan review robot during the monitoring period, where i is the number of reviews of the plan review robot; Traverse the review status values ​​obtained by the review of all review files by the planned review robot within the monitoring period to obtain the maximum review status value Cmax and the minimum review status value Cmin; If the maximum review state value Cmax ≤ the review state comparison value C0 within the monitoring period, a level 4 state signal of the plan review robot is generated; If the minimum review state value Cmin within the monitoring period is greater than the review state comparison value C0, a primary state signal of the plan review robot is generated; If the maximum review state value Cmax within the monitoring period is greater than the review state comparison value C0 and greater than the minimum review state value Cmin within the monitoring period, a first-level verification state signal of the plan review robot is generated.

3. The intelligent management method of plan review robot based on multi-source data analysis according to claim 2 is characterized in that: Based on the primary verification status signal of the plan review robot, the review status variance value of the plan review robot within the monitoring period is obtained, which is recorded as Cf; When the review state variance value Cf ≥ the review state variance threshold Cy, a secondary state signal of the plan review robot is generated; When the review state variance value Cf is less than the review state variance threshold Cy, a secondary verification state signal of the plan review robot is generated.

4. The intelligent management method of plan review robot based on multi-source data analysis according to claim 3 is characterized in that: Based on the secondary verification status signal of the plan review robot, the review status value of the plan review robot within the monitoring period is identified; If the review state value Ci of the plan review robot decreases linearly and continuously during the monitoring period, a third-level state signal of the plan review robot is generated; If the review state value Ci of the plan review robot increases linearly and continuously during the monitoring period, a primary state signal of the plan review robot is generated.

5. The intelligent management method of plan review robot based on multi-source data analysis according to claim 1 is characterized in that: During the monitoring cycle, hardware status monitoring and identification include: Obtain the electronic components that the planned review robot needs to monitor when reviewing documents; Record the items that need to be monitored for electronic components as monitoring items; The monitoring items in the electronic components whose monitoring values ​​do not meet the monitoring value requirements of the monitoring items are recorded as risk items; The risk items are processed to obtain the total risk value and risk item ratio of the electronic components; The total risk value of the electronic components is multiplied by the risk item ratio to obtain the risk screening value of the electronic components; The risk screening values ​​obtained each time the documents of the same electronic component are reviewed are summed and averaged to obtain the risk screening analysis value of the electronic component during the monitoring period.

6. The intelligent management method of plan review robot based on multi-source data analysis according to claim 5 is characterized in that: The risk analysis values ​​of all electronic components that need to be monitored in the plan review robot are sorted out, and the risk analysis values ​​are sorted in descending order to obtain the risk level of the corresponding electronic components and conduct screening and processing.

7. The intelligent management system of plan review robot based on multi-source data analysis is characterized by: include: A data analysis module, which is used to obtain the review status data of the plan review robot during the review process and transmit the review status data to the cloud management and control platform; A status verification module, wherein the status verification module receives the review status data transmitted by the cloud management and control platform, processes the review status data, obtains the review status value of the planned review robot, and sends the review status value to the cloud management and control platform; A multimodal analysis module, wherein the multimodal analysis module receives the review status value transmitted by the cloud management and control platform. During the monitoring period, the multimodal analysis module analyzes and processes the review status value of the plan review robot and the review status comparison value of the plan review robot to obtain the review status signal of the plan review machine during the monitoring period, and transmits the review status signal to the cloud management and control platform; The review status signal includes a primary status signal, a secondary status signal, a tertiary status signal and a quaternary status signal of the planned review robot; A performance control module, which receives the review status signal transmitted by the cloud control platform. The performance control module monitors and identifies the hardware status of the planned review robot within the monitoring period, and completes the visual identification of hardware control; The process of obtaining the review status value of the planned review robot is as follows: The censored external analysis value is recorded as Cw; The internal factor analysis value of the review is recorded as Cn; The external disturbance value is recorded as Cr; By formula The review state value Ci of the plan review robot is calculated, where Si is the efficiency state value of the plan review robot; The process of obtaining the external analysis value is as follows: Get the time from when the planned review robot starts reviewing the review document to when the review report is generated, and record it as the total review time; Calculate the ratio of the total review time to the size of the review file to obtain the review external analysis value of the planned review robot; The process of obtaining the internal factor analysis value is as follows: Obtain the parsing speed and compression speed of the review file by the plan review robot, sum and average the parsing speed and compression speed of the review file by the plan review robot, and obtain the internal factor analysis value of the review file; The process of obtaining the external disturbance value is as follows: The total review time is divided into several time sub-units, and the network delay time of the planned review robot when reviewing the review file is obtained through the Ping test in each time sub-unit; Sum up the network delay times of all time subunits to obtain the total review delay time; The total review delay time is calculated by the ratio of the review file size to obtain the review out-of-review disturbance value of the review file; The process of obtaining the effectiveness status value of the plan review robot is as follows: Get the capacity effectiveness ratio of the plan review robot, denoted as Sr; Get the maximum throughput of the plan review robot, denoted as St; Get the historical packet loss rate of the plan review robot, denoted as Sd; By formula The efficiency state value Si of the plan review robot is calculated, where a1 and a2 are preset proportional coefficients.

Images