Power grid equipment risk monitoring system based on RPA

Abstract

The present invention discloses a risk monitoring system for power grid equipment based on RPA, and relates to the technical field of automatic identification. The present invention sets an automatic execution unit to periodically obtain a number of stored monitoring values ​​from a corresponding branch system, and periodically performs model analysis on them. Based on the identification result data with a signal quantity of 1, a result analysis module obtains a number of intervention comparison tables, and at the same time, an execution intervention unit receives the intervention comparison table and generates an intervention judgment white list. After the execution acquisition data of a received characteristic parameter is obtained, it is first matched according to the intervention judgment white list. For matching situations, the execution intervention unit performs execution intervention on it, thereby avoiding the occurrence of untimely identification of risk types according to the established collection cycle, and can effectively improve the accuracy and timeliness of risk management and control of target equipment, and reduce the possibility of equipment failure.

Description

Technical Field

[0001] The present invention belongs to the technical field of automatic identification, and in particular is a power grid equipment risk monitoring system based on RPA. Background Art

[0002] With the continuous development of the power industry and the increasing scale of power grids, the safe and stable operation of power grid equipment is crucial to the reliability of the entire power system. In the process of analyzing and identifying power grid equipment, risk analysis models are currently widely used to identify risks of power grid equipment.

[0003] The existing risk identification methods for power grid equipment mainly rely on collecting monitoring values ​​from various branch systems within the power grid, and then inputting them into the risk analysis model for analysis. In order to ensure the accuracy of the monitoring data, Robotic Process Automation (RPA) technology is usually used. RPA can collect monitoring values ​​of power grid equipment from various branch systems within the power grid 24 hours a day. During the collection process, the corresponding collection interval is set based on each parameter, and RPA periodically collects the corresponding parameters.

[0004] However, this approach has obvious limitations. Due to the use of periodic data collection, when the monitoring value of a parameter is abnormal, the data cannot be input into the risk identification model in a timely and effective manner for risk identification. This may result in the inability to detect equipment risks in a timely manner for a period of time after the abnormal situation occurs, thereby increasing the possibility of equipment failure and posing a potential threat to the safe and stable operation of the power grid;

[0005] In order to solve the above problems, the present invention proposes a solution. Summary of the invention

[0006] The present invention aims to solve the problems raised in the above-mentioned background technology; to this end, the present invention proposes a power grid equipment risk monitoring system based on RPA.

[0007] The RPA-based power grid equipment risk monitoring system includes:

[0008] A risk identification module is used to periodically input a plurality of received execution acquisition data of a target device in a target power grid into a risk identification model to obtain corresponding identification result data, wherein the identification result data at least includes a signal quantity, the signal quantity is selected from a number 1 or 0, and when the signal quantity is 1, the corresponding identification result data also includes a risk type, a risk data segment, and a plurality of risk identification parameters;

[0009] A result analysis module is used to analyze all the stored recognition result data with a signal quantity of 1 according to a preset analysis step to obtain an intervention comparison table of the target device based on several risk types, and the execution intervention unit extracts corresponding concrete parameters from the obtained several intervention comparison tables, removes duplicates from all the extracted concrete parameters, and generates an intervention determination whitelist according to the result after removal of duplicates;

[0010] When the execution intervention unit receives the execution acquisition data of any characteristic parameter of the target device in an execution collection cycle, the characteristic parameter is first matched with the intervention determination whitelist. If the characteristic parameter is included in a plurality of pre-determination whitelists, an intervention determination result instruction corresponding to the characteristic parameter is generated according to the preset intervention determination rule. Otherwise, no processing is performed.

[0011] Every time the automation execution unit receives an intervention determination result instruction of a characteristic parameter, it generates intervention execution data of the characteristic parameter according to a preset intervention execution rule and transmits it to the risk identification module.

[0012] Furthermore, when the risk identification module receives the first transmitted intervention execution data, it starts timing and inputs all intervention execution data received within the preset intervention execution time into the risk identification model to obtain corresponding identification result data. If the signal quantity in the identification result data is 1, the management personnel will be notified immediately for processing, otherwise no processing will be performed.

[0013] Furthermore, it also includes a collection execution terminal for automatically acquiring data by periodically collecting several characteristic parameters of target equipment from several branch systems of the target power grid.

[0014] Furthermore, the acquisition execution terminal also includes an automation execution unit, which is equipped with a pre-set robot process automation program, and the robot process automation program is in a state of running at all times.

[0015] Compared with the prior art, the present invention has the following beneficial effects:

[0016] (1) The present invention sets an automatic execution unit to periodically establish a communication channel with a corresponding branch system according to the frequency and duration of the characteristic parameters corresponding to the target device to obtain a number of monitoring values ​​of the corresponding characteristic parameters stored therein to obtain corresponding execution acquisition data, sets a risk identification module to periodically obtain corresponding identification result data, and analyzes the identification result data based on the signal quantity of 1 by the result analysis module to obtain an intervention comparison table of a number of target devices based on a number of risk types, thereby ensuring the accuracy of the final risk management and control;

[0017] (2) After obtaining the intervention comparison table, the present invention generates an intervention judgment whitelist by the execution intervention unit. After the execution acquisition data of a received characteristic parameter is received, it is first matched according to the intervention judgment whitelist. For the matched cases, the execution intervention unit performs intervention and extracts the monitoring values ​​within a certain delay time. This method avoids the occurrence of untimely identification of risk types due to the established collection cycle. When the monitoring value of a certain characteristic parameter shows the characteristics of a certain risk type, several characteristic parameters in each risk type that cause the risk together with the characteristic parameter are collected and identified within the preset time limit. This can effectively improve the accuracy and timeliness of risk management and control of the target equipment and reduce the possibility of equipment failure. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a data relationship flow chart of the present invention;

[0019] Figure 2 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0020] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0021] See also Figure 1 , 2 ,This application provides a risk monitoring system for power grid equipment based on RPA, including a collection execution terminal, a risk identification module and a result analysis module;

[0022] The acquisition execution terminal is used to automatically collect monitoring data of several characteristic parameters of the target equipment from several branch systems of the target power grid, wherein the branch system refers to a series of digital systems designed in the operation and management process of the target power grid for data acquisition, processing, storage and transmission related to the target equipment;

[0023] A branch system correspondingly collects and stores monitoring values ​​of several characteristic parameters of target equipment in the target power grid in real time;

[0024] The collection execution terminal includes an automated execution unit and an execution intervention unit. The automated execution unit is equipped with a pre-set robotic process automation program. The robotic process automation program is used to periodically establish a communication channel with the corresponding branch system according to the frequency and duration of the corresponding characteristic parameters. After the communication channel is established, several monitoring values ​​of the corresponding characteristic parameters are obtained. In this embodiment, the robotic process automation program is in a state of running at all times.

[0025] In this embodiment, several characteristic parameters of the target device each correspond to a preset frequency duration;

[0026] For a characteristic parameter of the target device, the automated execution unit acquires all monitoring values ​​of the characteristic parameter stored in the execution collection cycle from the branch system that acquires the characteristic parameter at intervals of an execution collection cycle, generates execution acquisition data of the characteristic parameter in the execution collection cycle according to all the acquired monitoring values, and transmits the execution acquisition data of the characteristic parameter in the execution collection cycle to the risk identification module and the execution intervention unit respectively;

[0027] The risk identification module is used to periodically identify risks of target devices in the target power grid. The risk identification module pre-stores a trained risk identification model, and the risk identification model is used to identify risks of target devices in the target power grid;

[0028] For any characteristic parameter of the target device, the risk identification module stores the execution acquisition data of the characteristic parameter after receiving the execution acquisition data of the characteristic parameter in an execution acquisition cycle;

[0029] The risk identification module inputs all the execution acquisition data stored in the risk identification cycle into the risk identification model at each risk identification cycle, and the risk identification model identifies the risk and outputs the identification result data of the target device of the corresponding risk identification cycle. The identification result data at least includes a signal quantity. In this embodiment, the signal quantity is selected from the numbers 1 or 0. If the signal quantity is 1, it means that the corresponding risk identification cycle module identifies that the target device has a risk, otherwise it means that the corresponding risk identification cycle target device has not been identified as having a risk;

[0030] When the signal quantity is 1, the corresponding identification result data also includes the risk type, risk data segment and several risk identification parameters. The risk type refers to the type of risk that the risk identification model identifies as the target device. The risk data segment refers to several continuous monitoring values ​​corresponding to the risk type identified by the risk identification model. The risk identification model identifies the risk type of the target device through these several continuous monitoring values.

[0031] Risk identification parameters refer to characteristic parameters of several continuous monitoring values ​​corresponding to the risk type identified by the risk identification model, where continuous monitoring values ​​refer to the continuous collection time of the monitoring values;

[0032] The characteristic parameter to which a continuous monitoring value belongs refers to the characteristic parameter corresponding to the collection of the continuous monitoring value;

[0033] When the signal quantity contained in the identification result data of a risk identification cycle is 1, the identification result data of the risk identification cycle is transmitted to the result analysis module for storage;

[0034] The result analysis module is used to analyze the identification result data of several stored risk identification cycles, and the analysis steps are as follows:

[0035] S11: Select the target device as the device to be analyzed, obtain the identification result data of all risk identification cycles of the device to be analyzed stored in the result analysis module and mark them as A1, A2...Aa, a≥1;

[0036] S12: Traverse the identification result data A1, A2...Aa to obtain all risk types contained therein, remove duplicates from all the obtained risk types, and mark all remaining risk types after removal of duplicates as B1, B2...Bb, where b≥1;

[0037] S13: First, select the risk type B1 as the type to be analyzed, obtain all the identification result data containing the type to be analyzed from the identification result data A1, A2...Aa, and mark them as C1, C2...Cc, 1≤c <a;

[0038] S14: marking characteristic parameters of several continuous monitoring values ​​contained in the recognition result data C1 as E1, E2...Ee, e≥1;

[0039] S15: Calculate and obtain the recognition change characteristic value and time display characteristic value of the characteristic parameter E1 in the recognition result data C1 according to the preset calculation rules. The calculation rules are as follows:

[0040] S151: The continuous monitoring data corresponding to the characteristic parameter E1 in the recognition result data C1 are marked as F1, F2 ... Ff, f ≥ 1 in order of the acquisition time;

[0041] The monitoring difference G1 between the monitoring values ​​F1 and F2 is calculated using the formula G1=F2-F1;

[0042] S152: according to S151, the monitoring differences of the monitoring values ​​F2 and F3, F3 and F4, ... Ff-1 and Ff are calculated in sequence, and marked as G2, G3, ... Gg, where g=f-1;

[0043] S153: Compare the monitoring difference values ​​G1 and G. If G1≥G, select the collection time of the monitoring value with the smaller mark subscript from the two monitoring values ​​for calculating the monitoring difference value G1 as a time to be identified for the characteristic parameter E1, where G is a preset identification selected difference threshold. Otherwise, no processing is performed.

[0044] S154: According to S153, the monitoring difference values ​​G2, G3 ... Gg and G are compared in size, and all the moments to be identified of the characteristic parameter E1 are obtained based on the comparison results, and then all the moments to be identified are marked as H1, H2 ... Hh in chronological order, 1≤h≤g;

[0045] S155: sequentially match the to-be-recognized moments H2, H3, ... Hh-1 with the preset rejection conditions. There is a to-be-recognized moment Hi that satisfies the preset rejection condition: Hi-(Hi -1 )>H or Hi +1 -Hi>H, then in the time to be identified H1, H2, H3 ... Hh will be located before Hi all the time to be identified removed, i value is 2, 3 ... h-1;

[0046] When the time to be identified Hh-1 matches the preset elimination condition, all remaining time to be identified are obtained, and then all remaining time to be identified are re-marked as J1, J2...Jj in the order of H1, H2...Hh, 1≤j <h;

[0047] S156: sequentially obtain the monitoring values ​​corresponding to the time to be identified J1, J2...Jj at the collection time, re-mark them as K1, K2...Kj, and calculate the change value L1 from the time to be identified J1 to J2 using the formula L1=K2-K1;

[0048] S157: According to S156, the change values ​​L2, L3, ... Ll, l = j-1 are calculated in sequence to obtain the time to be identified J2 to J3, J3 to J4 ... Jj-1 to Jj;

[0049] S158: Using formula Calculate and obtain the deviation M1 of the change values ​​L1, L2...Ll, and compare M1 with M. In the formula, L is the average value of Lm, and M is the preset deviation elimination comparison amount;

[0050] If M1≥M, delete the corresponding Lm in descending order of |Lm - L| and calculate the deviation M1 of the remaining Lm. Compare M1 with M again until M1 < M. Obtain the average value of all the changing values participating in the calculation of M1 at this time, calibrate the average value as the recognition change eigenvalue of the characteristic parameter E1, and select the recognition moment farthest from the current moment from all the recognition moments corresponding to the changing values participating in the calculation of M1 at this time;

[0051] Calibrate the interval duration between the selected recognition moment and the acquisition moment corresponding to the monitoring value F1 as the time appearance eigenvalue of the characteristic parameter E1;

[0052] S16: Calculate and obtain the recognition change eigenvalue of the characteristic parameter E1 in the recognition result data C2, C3... Cc in sequence according to S13 to S15, combine it with the recognition change eigenvalue of the characteristic parameter E1 in the recognition result data C1, and calculate and obtain the concrete change value of the type to be analyzed based on the characteristic parameter E1 by using the summation and averaging formula;

[0053] Calculate and obtain the time appearance eigenvalue of the characteristic parameter E1 in the recognition result data C2, C3... Cc in sequence, combine it with the time appearance eigenvalue of the characteristic parameter E1 in the recognition result data C1, and calculate and obtain the concrete time value of the type to be analyzed based on the characteristic parameter E1 by using the summation and averaging formula;

[0054] S17: Calculate and obtain the concrete change values and concrete time values of the type to be analyzed based on the characteristic parameters E2, E3... Ee in sequence according to S15 to S16;

[0055] Select the characteristic parameter corresponding to the smallest concrete time value from the concrete time values of the type to be analyzed based on the characteristic parameters E1, E2... Ee as the concrete parameter of the type to be analyzed;

[0056] For all the characteristic parameters among E1, E2... Ee that are not selected as the concrete parameter, calculate the difference between their concrete time values and the concrete time value of the concrete parameter in sequence, and use the calculated result as the concrete delay time of the type to be analyzed based on the corresponding characteristic parameter;

[0057] Generate an intervention comparison table of the type to be executed according to the concrete parameter of the type to be analyzed and its concrete time value, and all the other characteristic parameters that are not selected as the concrete parameter and their concrete delay times;

[0058] The intervention comparison table contains a parameter field and a time field. In the parameter field, the concrete parameter and e - 1 characteristic parameters are stored in sequence from top to bottom, and in the time field, the concrete time value and e - 1 concrete delay times are stored correspondingly;

[0059] S18: selecting the risk types B2, B3, ... Bb to be analyzed in sequence, and generating intervention comparison tables of the risk types B2, B3, ... Bb in sequence according to S13 to S17;

[0060] The intervention comparison tables of risk types B1, B2...Bb are transmitted to the intervention execution unit. After receiving the transmitted intervention comparison tables of all risk types, the intervention execution unit extracts the concrete parameters contained in the intervention comparison tables of each risk type, removes duplicates from all the extracted concrete parameters, generates an intervention determination whitelist based on all the remaining concrete parameters after removal of duplicates, and stores the intervention determination whitelist;

[0061] After receiving the execution acquisition data of the characteristic parameters of the execution collection period, the execution intervention unit first matches it with the intervention determination whitelist. If the characteristic parameters are included in several pre-determination whitelists, the intervention determination data of the characteristic parameters of the execution collection period is generated according to the preset intervention determination rules. The intervention determination rules are as follows:

[0062] S21: marking all monitoring values ​​contained in the execution acquisition data as N1, N2, ..., Nn, in sequence according to the order of collection time, and calculating the difference between two adjacent monitoring values ​​in sequence, wherein two adjacent monitoring values ​​refer to the continuous collection time;

[0063] S22: If the absolute values ​​of q consecutive differences are all greater than R1, firstly, all intervention comparison tables whose concrete parameters are the characteristic parameters are obtained from the intervention execution unit and are marked as T1, T2...Tt, t≥1, R1 is a preset initial intervention comparison value based on the characteristic parameter, otherwise no processing is performed, and q is a preset initial intervention comparison value;

[0064] Then, according to the intervention comparison tables T1, T2...Tt, an intervention determination result instruction corresponding to a number of characteristic parameters of the intervention comparison table is generated according to a preset determination generation rule;

[0065] Taking the intervention comparison table T1 as an example, the determination generation rules of the intervention determination result instructions for generating several characteristic parameters of the intervention comparison table T1 are as follows:

[0066] SS31: Obtain the concrete delay time of all characteristic parameters from the intervention comparison table T1, and mark all characteristic parameters contained in the intervention comparison table T1 as U1, U2, ..., Uu in ascending order of the concrete delay time, where u ≥ 1, and mark the concrete delay time of the characteristic parameters U1, U2, ..., Uu in the intervention comparison table T1 as V1, V2, ..., Vu;

[0067] SS32: taking the time of receiving the execution acquisition data of the characteristic parameter of the execution acquisition cycle as the intervention determination identification time, and selecting whether to generate the intervention determination result instruction of the characteristic parameter U1 according to the preset selection generation rule;

[0068] The selected generation rule is as follows: starting from the intervention determination identification time, if the automation execution unit will obtain all the monitoring values ​​of the characteristic parameter U1 stored from the branch system that collects the characteristic parameter U1 according to the frequency duration of the characteristic parameter U1 within the next V1 time, no processing is performed; otherwise, the intervention determination result instruction of the characteristic parameter U1 is generated according to the intervention determination identification time, the characteristic parameter U1, and the concrete delay time V1;

[0069] Selecting whether to generate an intervention determination result instruction of the characteristic parameters U2, U3, ..., Uu according to the same preset selection generation rule;

[0070] Transmitting the generated intervention determination result instructions of all characteristic parameters to the automation execution unit;

[0071] S23: If the absolute values ​​of the consecutive q differences are not all greater than R1, no processing is performed;

[0072] After receiving the intervention determination result instructions of the characteristic parameters of all intervention comparison tables, the automatic execution unit performs intervention execution on the corresponding characteristic parameters according to the preset intervention execution rules;

[0073] After receiving an intervention determination result instruction of a characteristic parameter, the characteristic parameter is intervened and executed to generate intervention execution data of the characteristic parameter. The intervention execution rules are as follows:

[0074] S31: Get the intervention determination result instruction of the characteristic parameter, the intervention determination identification time, characteristic parameters and the concrete delay time W1;

[0075] S32: starting from the intervention determination identification moment, after the concrete delay time W1, all monitoring values ​​stored after the last execution collection cycle are obtained from the branch system for collecting the characteristic parameters, and the intervention execution data of the characteristic parameters are generated according to the monitoring values;

[0076] After the intervention execution data corresponding to the intervention judgment result instruction of all received characteristic parameters is generated, it is transmitted to the risk identification module for re-identification. The re-identification steps are as follows: the risk identification module inputs it into the risk identification model to obtain the corresponding identification result data. If the signal quantity in the identification result data is 1, the management personnel is notified immediately for processing, otherwise no processing is performed;

[0077] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The RPA-based power grid equipment risk monitoring system is characterized by: include: A risk identification module is used to periodically input a plurality of received execution acquisition data of a target device in a target power grid into a risk identification model to obtain corresponding identification result data, wherein the identification result data at least includes a signal quantity, the signal quantity is selected from a number 1 or 0, and when the signal quantity is 1, the corresponding identification result data also includes a risk type, a risk data segment, and a plurality of risk identification parameters; A result analysis module is used to analyze all the stored recognition result data with a signal quantity of 1 according to a preset analysis step to obtain an intervention comparison table of the target device based on several risk types, and the execution intervention unit extracts corresponding concrete parameters from the obtained several intervention comparison tables, removes duplicates from all the extracted concrete parameters, and generates an intervention determination whitelist according to the result after removal of duplicates; When the execution intervention unit receives the execution acquisition data of any characteristic parameter of the target device in an execution collection cycle, the characteristic parameter is first matched with the intervention determination whitelist. If the characteristic parameter is included in a plurality of pre-determination whitelists, an intervention determination result instruction corresponding to the characteristic parameter is generated according to the preset intervention determination rule. Otherwise, no processing is performed. Every time the automated execution unit receives an intervention determination result instruction of a characteristic parameter, it generates intervention execution data of the characteristic parameter according to a preset intervention execution rule and transmits it to the risk identification module for re-identification.

2. The RPA-based power grid equipment risk monitoring system according to claim 1 is characterized in that: When the risk identification module receives the first intervention execution data transmitted, it starts timing and inputs all intervention execution data received within the preset intervention execution time into the risk identification model to obtain the corresponding identification result data. If the signal quantity in the identification result data is 1, the management personnel will be notified immediately for processing, otherwise no processing will be performed.

3. The RPA-based power grid equipment risk monitoring system according to claim 1 is characterized in that: It also includes a collection execution terminal for automatically acquiring data by periodically collecting several characteristic parameters of target equipment from several branch systems of the target power grid.

4. The RPA-based power grid equipment risk monitoring system according to claim 3 is characterized in that: The automation execution unit is equipped with a pre-set robotic process automation program, which is in a state of running all the time.

5. The RPA-based power grid equipment risk monitoring system according to claim 1 is characterized in that: For a characteristic parameter of the target device, the automated execution unit obtains all monitoring values ​​of the characteristic parameter stored in the execution collection cycle from the branch system that collects the characteristic parameter at intervals of one execution collection cycle, and generates execution acquisition data of the characteristic parameter in the execution collection cycle based on all the acquired monitoring values.

6. The RPA-based power grid equipment risk monitoring system according to claim 1 is characterized in that: The analysis steps for obtaining the intervention comparison table of the target device based on several risk types are as follows: S11: Select the target device as the device to be analyzed, obtain the identification result data of all risk identification cycles of the device to be analyzed stored in the result analysis module and mark them as A1, A2...Aa, a≥1; S12: Traverse the identification result data A1, A2...Aa to obtain all risk types contained therein, remove duplicates from all the obtained risk types, and mark all remaining risk types after removal of duplicates as B1, B2...Bb, where b≥1; S13: First, select the risk type B1 as the type to be analyzed, obtain all the identification result data containing the type to be analyzed from the identification result data A1, A2...Aa, and mark them as C1, C2...Cc, 1≤c <a; S14: marking characteristic parameters of several continuous monitoring values ​​contained in the recognition result data C1 as E1, E2...Ee, e≥1; S15: Calculate and obtain the recognition change characteristic value and time display characteristic value of the characteristic parameter E1 in the recognition result data C1 according to the preset calculation rules; S16: according to S13 to S15, the recognition change characteristic value of the characteristic parameter E1 in the recognition result data C2, C3...Cc is calculated in sequence, and the recognition change characteristic value of the characteristic parameter E1 in the recognition result data C1 is combined, and the concrete change value of the type to be analyzed based on the characteristic parameter E1 is calculated by using the addition and averaging formula; The time display characteristic values ​​of the characteristic parameter E1 in the recognition result data C2, C3...Cc are calculated in sequence, and the time display characteristic values ​​of the characteristic parameter E1 in the recognition result data C1 are combined with the time display characteristic values ​​of the characteristic parameter E1 in the recognition result data C1, and the concrete time value of the type to be analyzed based on the characteristic parameter E1 is calculated by using the addition and averaging formula; S17: according to S15 to S16, the concrete change value and the concrete time value of the type to be analyzed based on the characteristic parameters E2, E3...Ee are calculated in sequence, and the characteristic parameter corresponding to the concrete time value with the smallest value is selected from the concrete time values ​​of the type to be analyzed based on the characteristic parameters E1, E2...Ee, and it is used as the concrete parameter of the type to be analyzed. Similarly, the concrete delay time of the type to be analyzed based on the corresponding characteristic parameter can be obtained; Generate an intervention comparison table for the type to be performed according to the concrete parameters of the type to be analyzed and their concrete time values, and all other characteristic parameters that are not selected as concrete parameters and their concrete delay times; S18: Select the risk types B2, B3, ... Bb to be analyzed in turn, and generate intervention comparison tables for the risk types B2, B3, ... Bb in turn according to S13 to S17.

7. The RPA-based power grid equipment risk monitoring system according to claim 1, characterized in that: The preset intervention determination rules for generating the intervention determination result instruction of the characteristic parameters of the execution acquisition cycle are as follows: S21: marking all monitoring values ​​contained in the execution acquisition data as N1, N2, ..., Nn, in sequence according to the order of collection time, and calculating the difference between two adjacent monitoring values ​​in sequence, wherein two adjacent monitoring values ​​refer to the continuous collection time; S22: If the absolute values ​​of q consecutive differences are all greater than R1, firstly, all intervention comparison tables whose concrete parameters are the characteristic parameters are obtained from the intervention execution unit and are marked as T1, T2...Tt, t≥1, R1 is a preset initial intervention comparison value based on the characteristic parameter, otherwise no processing is performed, and q is a preset initial intervention comparison value; Then, according to the intervention comparison tables T1, T2...Tt, an intervention determination result instruction corresponding to a number of characteristic parameters of the intervention comparison table is generated according to a preset determination generation rule; S23: If the absolute values ​​of q consecutive differences are not all greater than R1, no processing is performed.

8. The RPA-based power grid equipment risk monitoring system according to claim 7 is characterized in that: The determination generation rules of the intervention determination result instructions for generating several characteristic parameters according to the intervention comparison table T1 are as follows: SS31: Obtain the concrete delay time of all characteristic parameters from the intervention comparison table T1, and mark all characteristic parameters contained in the intervention comparison table T1 as U1, U2, ..., Uu in ascending order of the concrete delay time, where u ≥ 1, and mark the concrete delay time of the characteristic parameters U1, U2, ..., Uu in the intervention comparison table T1 as V1, V2, ..., Vu; SS32: The time when the execution acquisition data of the characteristic parameter of the execution collection cycle is received is used as the intervention determination identification time, and whether to generate the intervention determination result instruction of the characteristic parameter U1 is selected according to the preset selection generation rule.

9. The RPA-based power grid equipment risk monitoring system according to claim 8, characterized in that: SS32, the selection and generation rules for selecting whether to generate the intervention determination result instruction of the characteristic parameter U1 are as follows: Starting from the intervention judgment identification moment, if the automation execution unit will obtain all the stored monitoring values ​​of the characteristic parameter U1 from the branch system that collects the characteristic parameter U1 according to the frequency duration of the characteristic parameter U1 within the next V1 time, no processing will be done; otherwise, the intervention judgment result instruction of the characteristic parameter U1 is generated according to the intervention judgment identification moment, the characteristic parameter U1, and the concrete delay time V1.

10. The RPA-based power grid equipment risk monitoring system according to claim 1, characterized in that: The intervention execution rules for generating intervention execution data of a characteristic parameter based on the received intervention determination result instruction are as follows: S31: Get the intervention determination result instruction of the characteristic parameter, the intervention determination identification time, characteristic parameters and the concrete delay time W1; S32: starting from the intervention determination identification moment, after the concrete delay time W1, all monitoring values ​​stored after the last execution collection cycle are obtained from the branch system that collects the characteristic parameters, and the intervention execution data of the characteristic parameters are generated therefrom.

Images