An equipment whole life cycle management and control system based on an agent method

Abstract

The present application relates to the field of fault identification, and more particularly to a device full life cycle management and control system based on an agent method, comprising: determining whether to increase the review duration according to the input dynamic reference value and the input discrete reference value; determining to perform response timeliness analysis or parameter similarity analysis according to the strategy control density value and the input interval effective jump value; determining whether to adjust the preset response timeliness according to the effective analysis frequency and the early warning time superposition value; determining whether to analyze the granularity adaptation value according to the input form similarity; and determining whether the control strategy exists a control implicit fault according to the response timeliness and the granularity adaptation value. The present application improves the life management effect of the energy storage device through flexible analysis of the management strategy of the power grid energy storage device.

Description

Technical Field

[0001] This invention relates to the field of power safety management, and in particular to a device lifecycle management system based on an intelligent agent method. Background Technology

[0002] As an important carrier for integrating distributed energy resources (DER), microgrids have become a key component in building new power systems. The operating characteristics of microgrids are highly complex, posing severe challenges to system protection. For example, for multi-level power response regulation technology for energy storage in microgrids, the accuracy of the regulation strategy and its dynamic optimization capability are crucial to the safety of equipment lifespan. Therefore, how to effectively verify whether the regulation strategy meets the safety management requirements is a problem of great concern to those skilled in the art.

[0003] Chinese Patent Publication No. CN108288855B discloses a dynamic strategy simulation and optimization method based on microgrids, including: firstly, establishing a microgrid operation strategy library to store and manage control strategies of the microgrid under different operating environments; dividing the control time period during the operation of the microgrid system, and periodically selecting a set of candidate strategies for the next time period from the strategy library according to the various power sources and load conditions in the microgrid; downloading the selected strategy set to a microgrid simulation model built based on MATLAB, simulating the operation state of the microgrid, and selecting the optimal operation strategy as the operation strategy for the next time period by comparing and analyzing the simulation system results under each candidate strategy; comparing and analyzing the strategy operation results with the simulation calculation results, and using a genetic algorithm to improve and optimize the applied strategy. It can be seen that the above technical solution optimizes the applied strategy by comparing the strategy operation results with the simulation calculation results to achieve better results, but it has the following problems: the verification of the control strategy cannot be dynamically adapted to the scene characteristics, which easily leads to poor accuracy of the control strategy, and thus easily poses risks to the safety management of equipment lifespan. Summary of the Invention

[0004] Therefore, this invention provides a device lifecycle management system based on the intelligent agent method to overcome the problem in the prior art that the verification of power grid control strategies cannot be dynamically adapted to the characteristics of the scenario, which easily leads to poor accuracy of the control strategy and thus makes the safety management of equipment lifecycle risky.

[0005] To achieve the above objectives, this method provides a device lifecycle management system based on an intelligent agent method, comprising: The data analysis unit is used to determine the energy storage reference state based on the input dynamic reference value and the input discrete reference value, and to determine whether to increase the review time based on the energy storage reference state. The strategy analysis unit is used to determine the analysis conditions based on the strategy control density value and the effective jump value of the input interval, and to determine whether to perform response time analysis or execution parameter similarity analysis based on the analysis conditions. When performing response timeliness analysis, the first intelligent agent unit determines whether to adjust the preset response timeliness based on the effective analysis frequency and the superposition value of warning times. When performing parameter similarity analysis, the second intelligent agent unit determines whether to analyze the granularity adaptation value based on the input morphological similarity. The control and early warning unit determines the comparison conditions based on the timeliness of response and the granularity adaptation value, and determines whether to execute the life safety early warning based on the comparison conditions.

[0006] Furthermore, the data analysis unit determines to increase the review duration for energy storage reference states where the input dynamic reference value is greater than the preset input dynamic reference value or the input discrete reference value is greater than the preset input discrete reference value.

[0007] Furthermore, the data analysis unit adjusts the review duration based on the diversity of strategy control; The increase in the review duration is positively correlated with the diversity of strategy regulation.

[0008] Furthermore, the strategy analysis unit determines the execution response timeliness analysis based on the analysis conditions that the strategy control density value is greater than the preset strategy control density value and the effective jump value of the input interval is less than or equal to the preset effective jump value of the input interval.

[0009] Furthermore, for process states where the effective analysis frequency is greater than the preset effective analysis frequency and the cumulative value of warning times is greater than the preset cumulative value of warning times, the first intelligent agent unit determines to increase the preset response timeliness.

[0010] Furthermore, the first intelligent agent unit increases and adjusts the preset response timeliness based on the effective analysis frequency and the superposition value of the early warning time; The adjustment range of the preset response timeliness is positively correlated with the effective analysis frequency and the superposition value of the early warning time.

[0011] Furthermore, the strategy analysis unit determines the similarity analysis of execution parameters based on the analysis conditions that the strategy control density is less than or equal to the preset strategy control density and the effective jump value of the input interval is greater than the preset effective jump value of the input interval.

[0012] Furthermore, the second intelligent agent unit analyzes the granularity adaptation value based on the similarity condition that the similarity of the input shape is greater than the preset similarity of the input shape.

[0013] Furthermore, the second intelligent agent unit calculates the granularity adaptation value by weighted summation based on control smoothness and effect difference.

[0014] Furthermore, the control and early warning unit determines the execution life safety warning based on the comparison condition that the response timeliness is less than or equal to the preset response timeliness or the granularity adaptation value is less than or equal to the preset granularity adaptation value.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention determines the energy storage state by inputting dynamic reference values ​​and inputting discrete reference values, and determines whether to increase the review time based on the energy storage state. The input dynamic reference value reflects the degree of change of frequency deviation of the microgrid within the review time, and the input discrete reference value reflects the overall situation of frequency transformation of the microgrid. When the input dynamic reference value is greater than the preset input dynamic reference value or the input discrete reference value is greater than the preset input discrete reference value, the review time is increased to obtain more control records for analysis of the control strategy, making the acquisition of control records more flexible.

[0016] Furthermore, this invention determines the analysis conditions of the control record by using the strategy control density value and the effective jump value of the input interval. The strategy control density value reflects the temporal density of the control command issued by the control strategy, and the effective jump value of the input interval reflects the small and frequent fluctuations of the microgrid under the control strategy. This allows for the determination of whether to perform timeliness analysis or parameter similarity analysis, thus achieving precise determination of the analysis method and avoiding the inability of a single analysis method to meet the changing analysis conditions.

[0017] Furthermore, this invention determines whether to increase the preset response timeliness by combining the effective analysis frequency with the superimposed value of the warning time. The effective analysis frequency reflects the drastic change in the control commands of each target energy storage device within the review period, while the superimposed value of the warning time reflects the superposition of control records generated between the issuance of the control command and the completion of the control command by the control strategy. This further determines the adjustment range of the preset response timeliness, avoiding the problem that a fixed preset value cannot meet the changing scenario characteristics, thereby improving the analysis accuracy of the control strategy and further realizing predictive analysis of microgrid faults based on the accuracy of the control strategy.

[0018] Furthermore, this invention reflects the similarity of target energy storage devices in the control sequence by inputting morphological similarity, and under the similarity condition that the input morphological similarity is greater than the preset input morphological similarity, it determines whether to perform granular adaptation value analysis on the target energy storage devices in the analysis group based on the number of devices in the analysis group, thereby further improving the accuracy of the verification control strategy.

[0019] Furthermore, the present invention achieves response timeliness analysis and parameter similarity analysis through the first intelligent agent unit and the second intelligent agent unit, that is, it reduces the excessive input of human resources through software programs and improves the efficient utilization of resources. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the device lifecycle management system based on the intelligent agent method of the present invention; Figure 2 This is a flowchart illustrating how the present invention determines whether to increase the review duration based on input dynamic reference values ​​and input discrete reference values. Figure 3 This is a flowchart illustrating how the present invention determines whether to adjust the preset response timeliness based on the effective analysis frequency and the superimposed value of the early warning time. Figure 4 This is a flowchart illustrating the process of determining whether to implement a lifespan safety warning based on response timeliness and granularity adaptation value according to the present invention. Detailed Implementation

[0021] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0022] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0023] Please see Figures 1 to 4 As shown, the present invention provides a device lifecycle management system based on an intelligent agent method, comprising: The data analysis unit is used to determine the energy storage reference state based on the input dynamic reference value and the input discrete reference value, and to determine whether to increase the review time based on the energy storage reference state. The strategy analysis unit, which is connected to the data analysis unit, is used to determine the analysis conditions based on the strategy control density value and the effective jump value of the input interval, and to determine whether to perform response time analysis or parameter similarity analysis based on the analysis conditions. The first intelligent agent unit, which is connected to the strategy analysis unit, determines whether to adjust the preset response timeliness based on the effective analysis frequency and the superposition value of the early warning time when performing response timeliness analysis. The second intelligent agent unit, which is connected to the strategy analysis unit, determines whether to perform analysis on the granularity adaptation value based on the input morphological similarity when performing parameter similarity analysis. The control and early warning unit is connected to the first intelligent agent unit and the second intelligent agent unit respectively. It determines the comparison conditions based on the response timeliness and granularity adaptation value, and determines whether to execute the life safety early warning based on the comparison conditions.

[0024] In this invention, the control strategy is a multi-stage power response strategy for energy storage. The control strategy is executed by the energy storage system controller, which monitors the actual frequency of the microgrid's grid connection point and the preset rated frequency in real time to obtain the frequency deviation. And transmit it to the microgrid central controller, the microgrid central controller receives it and, according to... The threshold range corresponds to the generation of a control command. This control command reaches the energy storage system controller via the communication network. The control command is used to control the target energy storage device to charge or discharge power according to a specified adjustment range, until the energy storage system controller detects a new control command or the input or output power of the target energy storage device reaches the required condition of the control command (±0.2% of rated power), thus completing the current control. It can be understood that this invention presets several threshold ranges, and the control commands corresponding to different threshold ranges are different. This is already known to those skilled in the art and will not be elaborated here. In this embodiment of the invention, the threshold ranges and their corresponding control commands are as follows: ; Actual frequency - rated frequency. In this embodiment of the invention, the rated frequency is 50Hz. The target energy storage device is a device in the microgrid that can perform energy buffering and power regulation. The target energy storage device can be, but is not limited to, an energy storage cabinet.

[0025] This invention sets a review period for acquiring control records, which is used to acquire all control records within the review period and the corresponding time of the control records. The single review period is 24 hours. The longer the review period, the larger the amount of data processed when analyzing the control strategy, and the more accurate the analysis results. The more sensitive the user is to the accuracy of the analysis results, the longer the review period.

[0026] For a single target energy storage device, any control record includes the corresponding control strategy executed in a single instance. , The receiving time and the corresponding control commands, The reception time is the frequency deviation that the central controller receives from the energy storage system controller. At that moment.

[0027] Specifically, the data analysis unit determines to increase the review time for energy storage reference states where the input dynamic reference value is greater than the preset input dynamic reference value or the input discrete reference value is greater than the preset input discrete reference value.

[0028] Specifically, the control records corresponding to each target energy storage device within the most recent review period are extracted. For a single target energy storage device, the input dynamic reference value is the average of the input dynamic reference components for each target energy storage device. For a single target energy storage device, the input dynamic reference component is the number of analysis differences greater than a preset analysis difference. The analysis difference is the sum of the values ​​of two adjacent control records in the control sequence. The absolute value of the difference, the control sequence is the control record of the target energy storage device according to The sequence of receiving times arranged from earliest to latest. In this embodiment of the invention, the preset analysis difference value is the average value of all analysis differences corresponding to each target energy storage device. The larger the analysis difference value, the greater the frequency change of the microgrid, and the greater the change in the adjustment range of the target energy storage device. The higher the user's attention to the adjustment range of the target energy storage device, the smaller the value of the preset analysis difference value.

[0029] The input discrete reference value is the average of the input discrete reference components of each target energy storage device. For a single target energy storage device, the control sequence of the target energy storage device is obtained, and its input discrete reference component is the average of the analysis differences.

[0030] The method for confirming the diversity of policy control is to detect the control instructions corresponding to the policy control instructions within the review period. ,Will The number of corresponding threshold intervals is denoted as the strategy control diversity. The adjusted review duration = the review duration before adjustment × (strategy control diversity / preset strategy control diversity). It is worth noting that the review duration before adjustment is 24 hours. To avoid calculation redundancy caused by historical control records, the adjusted review duration has a maximum allowable value of 48 hours.

[0031] If the review duration is increased, the number of control records acquired for a single target energy storage device will be the same as the number of control records acquired within the increased review duration. The final acquired control records will be used as analysis data for the control strategy.

[0032] If the input dynamic reference value is less than or equal to the preset input dynamic reference value and the input discrete reference value is less than or equal to the preset input discrete reference value, then there is no need to increase the review time. For a single target energy storage device, the number of control records obtained is still the number of control records within the review time. The final obtained control records are used as the analysis data for the control strategy.

[0033] In this embodiment of the invention, there are 8 preset input dynamic reference values. It can be understood that the input dynamic reference values ​​reflect the degree of frequency change of the microgrid during the review period. The larger the input dynamic reference value, the more drastic the frequency change of the microgrid. The more users pay attention to the drastic degree of frequency change of the microgrid, the lower the preset input dynamic reference value.

[0034] In this embodiment of the invention, the preset input discrete reference value is 0.18Hz. It can be understood that the input discrete reference value reflects the overall situation of the microgrid's frequency transformation. The larger the input discrete reference value, the higher the frequency transformation amplitude of the microgrid. The more sensitive the user is to the frequency transformation of the microgrid, the lower the preset input discrete reference value.

[0035] In this embodiment of the invention, the preset strategy control diversity value is 5. It can be understood that the strategy control diversity reflects the degree of diversity of control commands issued by the control strategy to control the microgrid. The greater the strategy control diversity, the more types of control commands issued by the control strategy, the higher the difficulty of microgrid control, the higher the requirements for the control strategy capability, and the higher the user's requirements for the control strategy capability, the lower the preset strategy control diversity value.

[0036] Specifically, the review duration is increased based on the diversity of strategy control. The increase in the review duration is positively correlated with the diversity of strategy regulation.

[0037] Specifically, the strategy analysis unit determines the execution response timeliness analysis based on the analysis conditions that the strategy control density value is greater than the preset strategy control density value and the effective jump value of the input interval is less than or equal to the preset effective jump value of the input interval.

[0038] Response timeliness = the reciprocal of the time interval from when the control strategy issues a control command to when the target energy storage device begins to change its output power.

[0039] The control records corresponding to the microgrid within the review period are obtained. The strategy control density value = number of control records corresponding to the microgrid / review period. The effective jump value of the input interval is the average value of the effective jump components of the input interval for each target energy storage device. For a single target energy storage device, the method for confirming the effective jump component of its corresponding input interval is as follows: detect the control sequence corresponding to the target energy storage device and perform jump analysis on each control record to determine the effective jump record. The number of effective jump records is recorded as the effective jump component of the input interval for the target energy storage device. In the jump analysis of a single control record in the control sequence, it is checked whether it meets the preset comparison conditions with the comparison control record. If the preset comparison conditions are met, the control record is a valid jump record. Among them, the comparison control record is the control record that is sequentially adjacent to the control record. The preset comparison conditions are the two control records. Located in different threshold ranges but If the absolute value of the difference is less than or equal to 0.02Hz, in the jump analysis for a single target energy storage device, the last control record of the control sequence participates but is not subject to jump analysis.

[0040] In this embodiment of the invention, the preset strategy control density value is 6 times / h. It can be understood that the strategy control density value reflects the density of control commands issued by the control strategy in time. The larger the control strategy density value, the more intensive the control commands within the corresponding review period, the more frequent the frequency changes of the microgrid, the longer the time required for the control strategy to respond to these frequency changes, the higher the requirement for the timeliness of the control strategy response, and the higher the user's requirement for the density of control commands. Therefore, the preset strategy control density value is lower.

[0041] In this embodiment of the invention, the preset effective jump value of the input interval is 3. It can be understood that the effective jump value of the input interval reflects the small and frequent fluctuations of the microgrid in the control strategy. The larger the effective jump value of the input interval, the more serious the small and frequent fluctuations of the microgrid in the control strategy. Therefore, the analysis requirements for granularity adaptation value are higher. So, under the analysis conditions that the strategy control density is less than or equal to the preset strategy control density and the effective jump value of the input interval is greater than the preset effective jump value of the input interval, parameter similarity analysis is performed. Among them, the higher the user's requirements for the control strategy's ability to deal with frequent fluctuations, the lower the value of the preset effective jump value of the input interval.

[0042] Specifically, for process states where the effective analysis frequency is greater than the preset effective analysis frequency and the cumulative value of the early warning time is greater than the preset cumulative value of the early warning time, the first intelligent agent unit determines to increase the preset response timeliness.

[0043] For a single review duration, the effective analysis frequency is the sum of the number of control commands in each target energy storage device whose power jump amplitude is greater than the preset power jump amplitude. The control sequence of each target energy storage device is obtained. For a single control record of a single target energy storage device, the power jump amplitude is the absolute value of the difference between the adjustment amplitude of the control command corresponding to that control record and the adjustment amplitude of the control command corresponding to the next control record in the control sequence. In this embodiment, the absolute value of the difference between charging at 20% of the rated power and discharging at 20% of the rated power is 0.4, and the absolute value of the difference between charging at 20% of the rated power and charging at 50% of the rated power is 0.3. The preset power jump amplitude is the average value of the power jump amplitudes of each control command. The warning time superposition value is the maximum value of the warning superposition component of each control command. For a single control command, the warning superposition component is the number of control records generated within the warning interval corresponding to the control command. The warning interval is the time interval from the issuance of the control command to the final actual output power reaching the power corresponding to the control command.

[0044] In this embodiment of the invention, the preset effective analysis frequency is 18. It can be understood that the effective analysis frequency reflects the degree of overall adjustment of each target energy storage device within the review period. The higher the effective analysis frequency, the more drastic the change in the power adjustment amplitude of each target energy storage device. The higher the user's requirement for the ability of the control strategy to handle drastic changes in the control amplitude, the lower the value of the preset effective analysis frequency.

[0045] In this embodiment of the invention, the preset warning time superposition value is 2. It can be understood that the warning time superposition value reflects the superposition of control records generated between the time the control command is issued and the time the control strategy completes the control command. The larger the warning time superposition value, the more control records are generated within the warning interval corresponding to the control command, the greater the difficulty of microgrid control, and the higher the user's requirements for the control strategy's control capability of the microgrid. Therefore, the preset warning time superposition value is lower.

[0046] In this embodiment of the invention, the initial preset response timeliness is set to 3.5 / s. It can be understood that the response timeliness reflects the timeliness of the control strategy in resolving frequency events. The greater the response timeliness, the better the control effect of the control strategy on the microgrid, and the lower the possibility of hidden faults in the control strategy. Users are more sensitive to the possibility of hidden faults in the control strategy, and the higher the preset response timeliness value is.

[0047] Specifically, the first intelligent agent unit increases and adjusts the preset response timeliness based on the effective analysis frequency and the superposition value of the early warning time; The adjustment range of the preset response timeliness is positively correlated with the effective analysis frequency and the superposition value of the early warning time.

[0048] The adjusted preset response timeliness = the original preset response timeliness × adjustment coefficient. The adjustment coefficient is the smaller value between the frequency adjustment coefficient and the warning adjustment coefficient. The frequency adjustment coefficient = preset effective analysis frequency / effective analysis frequency, and the warning adjustment coefficient = preset warning time superposition value / warning time superposition value. It is worth noting that the adjusted preset response timeliness has a minimum value of 2 / s.

[0049] Specifically, the strategy analysis unit determines the similarity analysis of execution parameters based on the analysis conditions that the strategy control density is less than or equal to the preset strategy control density and the effective jump value of the input interval is greater than the preset effective jump value of the input interval.

[0050] Specifically, the second intelligent agent unit determines the granularity adaptation value for analysis based on the similarity condition that the input morphological similarity is greater than the preset input morphological similarity. Based on the input morphological similarity, a set of target device groups is constructed for the target energy storage devices. The target device group with the largest number of devices is designated as the analysis group. If the number of devices in the analysis group is greater than the preset number of devices, then the analysis is determined based on the granularity adaptation value.

[0051] The target device group is a collection of several target energy storage devices, and the input shape similarity between any two target energy storage devices in the target device group is greater than the preset input shape similarity. The similarity of input morphology between any two target energy storage devices is obtained by first acquiring the control sequences corresponding to the two target energy storage devices to obtain the similarity of input morphology between them. At the receiving time, if the difference in the number of control records is greater than or equal to the preset difference in the number of control records, the input morphological similarity is zero; if the difference in the number of control records is less than the preset difference in the number of control records, the maximum value of each amplitude difference is recorded as the input morphological similarity, for any given value in a single control sequence. Recorded as comparison The amplitude difference is the comparison. With another regulatory sequence target The absolute value of the difference, the target For comparison with another regulatory sequence The receiving time is closest in time to the control record. For a single target energy storage device, the difference in the number of control records is the absolute value of the difference in the number of control records corresponding to two target energy storage devices, and the preset difference in the number of control records is 5.

[0052] In this embodiment of the invention, the preset number of devices is 3. It can be understood that the larger the number of devices, the higher the similarity between the target energy storage devices, the higher the analysis accuracy of the effect difference of each target energy storage device when regulating the group to be analyzed, and the higher the user's requirements for the regulation accuracy of the regulation strategy, the higher the value of the preset number of devices.

[0053] In this embodiment of the invention, the preset input morphological similarity value is 0.2Hz. It can be understood that the input morphological similarity reflects the degree of similarity between each target energy storage device in the control sequence. The greater the input morphological similarity, the more similar the control records corresponding to each target energy storage device are, and the higher the value of similarity analysis of the target energy storage devices. The more sensitive the user is to the value of similarity analysis, the greater the preset input morphological similarity value.

[0054] Specifically, the second intelligent agent unit calculates the granularity adaptation value by weighted summation based on control smoothness and effect difference.

[0055] Particle size fit value ×Control smoothness / Preset control smoothness+ ×Preset effect difference / Effect difference, in this embodiment of the invention The value is 0.4. The value is 0.6.

[0056] The effect difference is the maximum value of the effect difference sub-values ​​corresponding to each threshold interval. For a single threshold interval, the effect difference sub-value is the maximum value of each threshold interval. The range of the corresponding effect reference value, for a single The calculation method for the effect reference value is as follows: ; The actual output amplitude is the value recorded in the microgrid control record. The actual output power of the microgrid at the receiving moment.

[0057] The control smoothness is the average value of the control smoothing components for each target energy storage device. For a single target energy storage device, the control smoothing components are calculated as follows: ; in, For the first in the regulatory sequence Each control command corresponds to a different output power. Let L be the output power corresponding to the q-th control command in the control sequence, and L be the number of control commands. To preset the power jump amplitude, This represents the amplitude of a single power transition.

[0058] In this embodiment of the invention, the preset control smoothness value is 100s. It can be understood that the control smoothness reflects the sensitivity of the diagnostic system to identifying different frequency changes in the microgrid. The greater the control smoothness, the more sensitive the diagnostic system is to the frequency changes of the microgrid, the finer the control granularity of the control strategy on the microgrid, and the smoother the control effect. The more sensitive the user is to the smoothness of the control effect of the control strategy, the higher the preset control smoothness value.

[0059] In this embodiment of the invention, the preset effect difference degree is set to 0.1. It can be understood that the effect difference degree reflects the degree of difference in the control effect of the control strategy on each target energy storage device. The higher the effect difference degree, the more different the control effect of the control strategy on each target energy storage device, the worse the adaptation conditions of the control strategy, and the more sensitive the user is to the adaptation conditions of the control strategy. Therefore, the preset effect difference degree is smaller.

[0060] Specifically, the control and early warning unit determines to execute a lifetime safety early warning based on comparison conditions where the response timeliness is less than or equal to a preset response timeliness or the granularity adaptation value is less than or equal to a preset granularity adaptation value. The lifetime safety early warning involves sending the corresponding control records of the microgrid within the review period to the user terminal to trigger an alarm. This helps users quickly obtain and locate control records with safety risks. Users can then target the corresponding control records... , The receiving time and corresponding control commands are checked to determine whether the control strategy has an impact on equipment safety.

[0061] Specifically, based on the comparison conditions that the response timeliness is greater than the preset response timeliness and the granularity adaptation value is greater than the preset granularity adaptation value, it is determined that there is no need to execute the life safety warning.

[0062] In this embodiment of the invention, the preset granularity adaptation value is set to 1. It can be understood that the granularity adaptation value reflects the degree of adaptation of the control strategy to the microgrid. The larger the granularity adaptation value, the higher the degree of adaptation of the control strategy to the microgrid, the better the control effect of the control strategy, and the higher the user's requirements for the control effect of the microgrid. Therefore, the preset granularity adaptation value is set to be higher.

[0063] Furthermore, this invention also provides a method for determining preset values. These preset values ​​include preset response timeliness, preset input dynamic reference value, preset input discrete reference value, preset analysis difference, preset strategy control diversity, preset strategy control density value, preset effective jump value for input intervals, preset effective analysis frequency, preset warning time superposition value, preset power jump amplitude, preset number of devices, preset input shape similarity, preset difference in the number of control records, preset control smoothness, and preset effect difference. Users can set preset values ​​not only based on their own experience but also by using historical data. For example, for preset response timeliness, extracting... The microgrid's control strategy utilizes several response times throughout its historical usage. The response times that meet user needs are selected from those historical processes. The average response time after removing outliers is recorded as the preset response time. In this invention, the control effect of the control strategy is used as the benchmark for determining whether the control strategy meets user needs. Those skilled in the art will know that users can choose indicators to quantify the processing capability of the control strategy, such as the number of occurrences of latent faults, which can also be used as benchmarks for determining whether the control strategy meets user needs. These will not be elaborated upon here. Methods for removing outliers include, but are not limited to, Z-Score or IQR.

[0064] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. An agent method-based equipment full life cycle management system, characterized in that, include: The data analysis unit is used to determine the energy storage reference state based on the input dynamic reference value and the input discrete reference value, and to determine whether to increase the review time based on the energy storage reference state. The strategy analysis unit is used to determine the analysis conditions based on the strategy control density value and the effective jump value of the input interval, and to determine whether to perform response time analysis or execution parameter similarity analysis based on the analysis conditions. When performing response timeliness analysis, the first intelligent agent unit determines whether to adjust the preset response timeliness based on the effective analysis frequency and the superposition value of warning times. When performing parameter similarity analysis, the second intelligent agent unit determines whether to analyze the granularity adaptation value based on the input morphological similarity. The control and early warning unit determines the comparison conditions based on the timeliness of response and the granularity adaptation value, and determines whether to execute the life safety early warning based on the comparison conditions.

2. The equipment lifecycle management system based on the intelligent agent method according to claim 1, characterized in that, The data analysis unit determines to increase the review time for energy storage reference states where the input dynamic reference value is greater than the preset input dynamic reference value or the input discrete reference value is greater than the preset input discrete reference value.

3. The equipment lifecycle management system based on the intelligent agent method according to claim 2, characterized in that, The data analysis unit adjusts the review duration based on the diversity of strategy control. The increase in the review duration is positively correlated with the diversity of strategy regulation.

4. The equipment lifecycle management system based on the intelligent agent method according to claim 2, characterized in that, The strategy analysis unit determines the execution response timeliness analysis based on the analysis conditions that the strategy control density value is greater than the preset strategy control density value and the effective jump value of the input interval is less than or equal to the preset effective jump value of the input interval.

5. The equipment lifecycle management system based on the intelligent agent method according to claim 4, characterized in that, For a process state where the effective analysis frequency is greater than the preset effective analysis frequency and the cumulative value of the early warning time is greater than the preset cumulative value of the early warning time, the first intelligent agent unit determines to increase the preset response timeliness.

6. The equipment lifecycle management system based on the intelligent agent method according to claim 5, characterized in that, The first intelligent agent unit increases and adjusts the preset response timeliness based on the effective analysis frequency and the superposition value of the early warning time; The adjustment range of the preset response timeliness is positively correlated with the effective analysis frequency and the superposition value of the early warning time.

7. The equipment lifecycle management system based on the intelligent agent method according to claim 2, characterized in that, The strategy analysis unit determines the similarity analysis of execution parameters based on the analysis conditions that the strategy control density is less than or equal to the preset strategy control density and the effective jump value of the input interval is greater than the preset effective jump value of the input interval.

8. The equipment lifecycle management system based on the intelligent agent method according to claim 7, characterized in that, The second intelligent agent unit determines the granularity adaptation value for analysis based on the similarity condition where the similarity of the input shape is greater than the preset similarity of the input shape.

9. The equipment lifecycle management system based on the intelligent agent method according to claim 8, characterized in that, The second agent unit calculates the granularity adaptation value by weighted summation based on control smoothness and effect difference.

10. The equipment lifecycle management system based on the intelligent agent method according to claim 9, characterized in that, The control and early warning unit determines the execution life safety warning based on the comparison condition that the response timeliness is less than or equal to the preset response timeliness or the granularity adaptation value is less than or equal to the preset granularity adaptation value.

Images