A power grid variable monitoring system and method

Abstract

The application discloses a kind of power grid variable monitoring system and method, it is related to power grid variable monitoring technical field.The system of the present application includes: data acquisition and processing module, feature extraction and analysis module, health assessment and risk judgment module and real-time monitoring and early warning module;Data acquisition and processing module constructs historical operation data set, feature extraction and analysis module identifies power grid abnormal fluctuation, health assessment and risk judgment module assesses energy storage equipment health and risk and establishes early warning reference, real-time monitoring and early warning module refers to historical analysis real-time data, matches early warning reference and generates prompt information.The present application not only can be able to real-time monitoring the operating state of energy storage equipment, also should be able to identify the potential risk of power grid operation by analyzing historical data, and then provide decision support for the dispatching and management of power grid, ensure the safe, stable operation of power grid.

Description

Technical Field

[0001] This invention relates to the field of power grid variable monitoring technology, specifically to a power grid variable monitoring system and method. Background Technology

[0002] As power systems continue to expand in scale and increase in complexity, the stability and reliability of power grids have become one of the core challenges in power dispatch and management. To ensure the efficient and stable operation of the power grid, real-time monitoring and analysis of various operational data are essential. Traditional power grid monitoring methods largely rely on static data and basic operational parameter monitoring, such as load, frequency, and voltage. While these methods can guarantee the normal operation of the power grid to a certain extent, with the increasing complexity of the power grid, traditional monitoring methods are no longer sufficient to effectively cope with sudden power grid anomalies and operational risks.

[0003] During power grid operation, energy storage devices play an increasingly important role as a crucial tool for regulating grid load and balancing power supply and demand. However, changes in the state of energy storage devices during grid operation, such as abnormal fluctuations in the charging and discharging process, can lead to load and frequency fluctuations in the grid, and may even trigger more serious grid faults. Therefore, the health status of energy storage devices and its relationship with abnormal grid fluctuations require more in-depth analysis.

[0004] Current technologies for monitoring energy storage devices and grid fluctuations primarily focus on the real-time acquisition and monitoring of individual data points, lacking a comprehensive analysis of the overall grid operating trends and the status of energy storage devices. For example, traditional methods often only focus on real-time fluctuations in grid load or frequency, neglecting the role of energy storage devices and the potential impact of their health status on grid operation. Furthermore, existing grid monitoring systems lack in-depth correlation analysis between historical dispatch commands and actual operating data, failing to accurately identify the intrinsic link between the risk status of energy storage devices and abnormal grid fluctuations. Summary of the Invention

[0005] The purpose of this invention is to provide a power grid variable monitoring system and method to solve the problems mentioned in the background art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] A method for monitoring power grid variables includes the following steps:

[0008] Step S100. Collect historical operation data and historical dispatch command records of the power grid in the past time period, analyze the historical dispatch command records, extract the historical dispatch commands for charging and discharging of energy storage devices, summarize and record them as the target command set; match the target command set with the corresponding historical operation data one-to-one to construct the historical operation dataset;

[0009] Step S200. For each element in the historical operation dataset, extract the corresponding key trend features and construct a key trend feature vector; based on the key trend feature vector, perform load fluctuation analysis and frequency fluctuation analysis of the power grid respectively, so as to identify abnormal fluctuations;

[0010] Step S300. Based on the key trend feature vector, assess the health status of the energy storage device corresponding to each target instruction; based on the assessment results of the health status of the energy storage device, determine whether the energy storage device is in a risky state of overcharging or over-discharging; if a risky state is determined, mark the corresponding target instruction as a risk instruction, obtain the correlation pattern between the risk instruction and abnormal fluctuations in the power grid, and mark it as a risk warning reference;

[0011] Step S400. Collect real-time operation data of the power grid, analyze the real-time operation data with reference to the analysis method of historical operation data, and obtain the risk assessment result of the current energy storage device; compare and analyze the risk assessment result of the current energy storage device with the corresponding historical operation dataset, thereby matching risk warning reference and generating corresponding prompt information.

[0012] Furthermore, step S100 includes:

[0013] S101. Collect historical operating data and historical dispatch command records of the power grid within a past time period. The historical operating data includes power grid load data, power grid frequency data, and energy storage system status data. The energy storage system status data includes the state of charge and charging / discharging power of the energy storage devices. From the historical dispatch command records, based on pre-set filtering keywords, select historical dispatch commands for charging and discharging of the energy storage devices and mark them as target commands. Summarize all target commands to form a target command set, represented as {I1,I2,...,In}, where I1 represents the first target command, I2 represents the second target command, and so on, In represents the nth target command, and n represents the total number of target commands within the past time period.

[0014] S102. Based on the call timestamp td of each target instruction in the target instruction set, find the historical execution data within the time period [td-T, td] and form a data pair (Ii, Di), where T represents the time step and Di represents the historical execution data corresponding to the i-th target instruction, i ranging from 1 to n; summarize all data pairs to construct a historical execution dataset S, which is represented as: S={(I1,D1),(I2,D2),...,(In,Dn)}, where D1 represents the historical execution data corresponding to the 1st target instruction, D2 represents the historical execution data corresponding to the 2nd target instruction, and so on, with Dn representing the historical execution data corresponding to the nth target instruction.

[0015] Furthermore, step S200 includes:

[0016] S201. For each element in the historical operation dataset, extract the corresponding key trend features; the key trend features include load fluctuation trend features, frequency fluctuation trend features, and energy storage device status trend features; wherein the load fluctuation trend feature is the dynamic change rate of load power ΔP between adjacent time points, and the corresponding calculation formula is: ΔP=[P_load(t+Δt)-P_load(t)] / Δt, where P_load(t+Δt) represents the grid load power at time (t+Δt), P_load(t) represents the grid load power at time t, and Δt represents the time interval; similarly, the frequency fluctuation trend feature is the frequency change rate, and the frequency change rate Δfc is calculated with reference to the calculation formula of the dynamic change rate of load power ΔP; the energy storage device status trend features include the energy storage charge / discharge power ratio Pc and the state of charge change rate. S, the corresponding calculation formulas are as follows: Pc=P_charge(t) / [P_discharge(t)+∈], where P_charge(t) represents the charging power of the energy storage device at time t, P_discharge(t) represents the discharging power of the energy storage device at time t; ∈ represents the minimum value, which is greater than 0; S=[SOC(t+Δt)-SOC(t)] / Δt, where SOC(t+Δt) represents the state of charge of the energy storage device at time (t+Δt), and SOC(t) represents the state of charge of the energy storage device at time t; the extracted key trend features are averaged and normalized within the time period [td-T, td], so as to construct a key feature vector Gi for each target instruction Ii, and Gi=[ΔP_load_i,Δf_i,P_i,SCR_i];

[0017] S202. Based on the key feature vector, statistical analysis is performed on the dynamic change rate of load power ΔP_load and the frequency change rate Δf to calculate the thresholds at different confidence levels, which are respectively the upper thresholds Tp1 and Tf1, and the lower thresholds Tp2 and Tf2. For load fluctuation anomaly analysis, ΔP_load_i in the key feature vector Gi is compared with the upper threshold Tp1 and the lower threshold Tp2. If ΔP_load_i > Tp1 or ΔP_load_i < Tp2, it is preliminarily determined that there is an abnormal load fluctuation. The corresponding energy storage charge / discharge power ratio P_i and the rate of change of state of charge SCR_i are used for auxiliary judgment. When it is initially determined that there is an abnormal load fluctuation, the abnormal load fluctuation is confirmed if the following conditions are met: If ΔP_load_i > Tp1, and P_i < 1 and SCR_i < 0, it means that the energy storage device may not have responded to the increase in load in time to discharge, thus confirming the abnormal load fluctuation; if ΔP_load_i < Tp2, and P_i > 1 and SCR_i > 0, it means that the energy storage device may not have effectively utilized the excess electrical energy, thus confirming the abnormal load fluctuation.

[0018] S203. For frequency fluctuation anomaly analysis, the frequency change rate Δf_i in the key feature vector Gi is compared with the upper threshold Tf1 and the lower threshold Tf2 respectively. If Δf_i > Tf1 or Δf_i < Tf2, then a frequency fluctuation anomaly is initially present. The Pearson correlation coefficient r(Δf_i, ΔP_load_i) between the frequency change rate Δf_i and the load power dynamic change rate ΔP_load_i is calculated. If |r(Δf_i, ΔP_load_i)| > r, and a frequency fluctuation anomaly is initially present, further confirmation is made based on the direction of load and frequency changes: when ΔP_load_i > 0 and Δf_i < 0, a frequency fluctuation anomaly is confirmed; when ΔP_load_i < 0 and Δf_i > 0, a frequency fluctuation anomaly is confirmed.

[0019] Furthermore, step S300 includes:

[0020] S301. Extract the energy storage charge / discharge power ratio and state of charge change rate from the key trend feature vector, assess the health of the energy storage device corresponding to each target instruction, and obtain the comprehensive health assessment score Hi. The corresponding calculation formula is as follows:

[0021] Hi=w1·P_i+w2·SCR_i+w3·σ(P_i,SCR_i),

[0022] Where σ(P_i,SCR_i) represents the standard deviation of the charge / discharge power ratio and the rate of change of state of charge of the energy storage device, w1, w2, and w3 represent weighting coefficients, and w1+w2+w3=1; the comprehensive health assessment score Hi is compared with the threshold interval Q, and Q=[H_min,H_max], where H_min represents the minimum value of the threshold interval Q, and H_max represents the maximum value of the threshold interval Q; if the comprehensive health assessment score Hi is not within the threshold interval Q, and P_i>1 and SCR_i>0, it is judged as an overcharging risk; if the comprehensive health assessment score Hi is not within the threshold interval Q, and P_i<1 and SCR_i<0, it is judged as an over-discharging risk;

[0023] S302. Based on the judgment result, the corresponding target instruction is marked as a risk instruction; according to the identification result of abnormal fluctuation in step S200, the abnormal fluctuation timestamp tb is obtained, and the abnormal fluctuation timestamp tb is associated with the time period of the historical operation data corresponding to the risk instruction, so as to obtain the association pattern between the risk instruction and the abnormal fluctuation of the power grid, and the association pattern between the risk instruction and the abnormal fluctuation of the power grid is marked as a risk warning reference.

[0024] The correlation pattern refers to the relationship between risk commands and abnormal fluctuations in the power grid. This is achieved by associating the timestamps of abnormal fluctuations with corresponding historical operational data time periods to form a pattern. This correlation pattern helps the system identify past anomalies and trigger warnings under similar circumstances. For example, if energy storage devices pose a risk of overcharging or discharging within a specific time period, and the power grid experiences abnormal fluctuations (such as voltage or frequency fluctuations) during that period, this time period can serve as a risk warning reference. When similar abnormal power grid fluctuations occur, the system will issue corresponding warnings based on this correlation pattern.

[0025] Furthermore, step S400 includes:

[0026] S401. Collect real-time operation data of the power grid, analyze the real-time operation data with reference to the analysis method of historical operation data, and extract the real-time key feature vector SG, where SG = [ΔPs_load, Δfs, Ps, SCRs], where ΔPs_load represents the real-time load fluctuation trend feature, Δfs represents the real-time frequency fluctuation trend feature, and Ps and SCRs represent the real-time status trend feature of the energy storage device; based on the real-time key feature vector SG, calculate the real-time health assessment score SH with reference to the calculation formula of the comprehensive health assessment score Hi; compare the real-time health assessment score SH with the threshold interval Q. If the real-time health assessment score SH is not within the threshold interval Q, and Ps > 1 and SCRs > 0, it is judged as an overcharging risk; if the comprehensive health assessment score Hi is not within the threshold interval Q, and Ps < 1 and SCRs < 0, it is judged as an over-discharging risk.

[0027] S402. Based on the risk assessment results, extract historical operational data of the same risk type from the risk instructions, calculate the similarity between the real-time operational data and the historical operational data in turn, and select the element with the highest similarity and the corresponding similarity value greater than the similarity threshold as the matching result; obtain the association pattern between the corresponding risk instructions and abnormal fluctuations in the power grid based on the matching results, and mark it as a risk warning reference; generate corresponding prompt information for relevant personnel based on the risk warning reference, and have the relevant personnel further handle the corresponding matters.

[0028] A power grid variable monitoring system includes: a data acquisition and processing module, a feature extraction and analysis module, a health assessment and risk judgment module, and a real-time monitoring and early warning module;

[0029] The data acquisition and processing module collects historical operation data and historical dispatch command records of the power grid over past time periods, analyzes the historical dispatch command records, extracts historical dispatch commands for charging and discharging of energy storage devices, summarizes and records them as a target command set; and establishes a one-to-one correspondence between the target command set and the corresponding historical operation data to construct a historical operation dataset.

[0030] The feature extraction and analysis module extracts the corresponding key trend features for each element in the historical operation dataset and constructs a key trend feature vector. Based on the key trend feature vector, load fluctuation analysis and frequency fluctuation analysis of the power grid are performed respectively to identify abnormal fluctuations.

[0031] The health assessment and risk judgment module evaluates the health status of the energy storage device corresponding to each target instruction based on the key trend feature vector; based on the assessment results of the energy storage device's health status, it determines whether the energy storage device is in a risky state of overcharging or over-discharging; if a risky state is determined, the corresponding target instruction is marked as a risk instruction, the correlation pattern between the risk instruction and abnormal fluctuations in the power grid is obtained, and it is marked as a risk warning reference.

[0032] The real-time monitoring and early warning module collects real-time operation data of the power grid, analyzes the real-time operation data with reference to the analysis method of historical operation data, and obtains the risk assessment result of the current energy storage equipment. The risk assessment result of the current energy storage equipment is compared and analyzed with the corresponding historical operation dataset to match risk warning references and generate corresponding prompt information.

[0033] Furthermore, the data acquisition and processing module includes a data acquisition unit and a data processing unit;

[0034] The data acquisition unit collects historical operation data and historical dispatch command records of the power grid over the past period; the data processing unit analyzes the historical dispatch command records, extracts the historical dispatch commands for charging and discharging of energy storage devices, summarizes and records them as the target command set; and maps the target command set to the corresponding historical operation data to construct a historical operation dataset.

[0035] Furthermore, the feature extraction and analysis module includes a feature extraction unit and an abnormal fluctuation analysis unit;

[0036] The feature extraction unit extracts the corresponding key trend features for each element in the historical operation dataset and constructs a key trend feature vector; the abnormal fluctuation analysis unit performs load fluctuation analysis and frequency fluctuation analysis of the power grid based on the key trend feature vector, thereby identifying abnormal fluctuations.

[0037] Furthermore, the health assessment and risk judgment module includes a health assessment unit, a risk judgment unit, and a risk instruction marking and association unit;

[0038] The health assessment unit evaluates the health status of the energy storage device corresponding to each target instruction based on the key trend feature vector; the risk judgment unit determines whether the energy storage device is in a risky state of overcharging or over-discharging based on the assessment results of the energy storage device's health status; if the risk instruction marking and association unit determines that a risky state exists, it marks the corresponding target instruction as a risk instruction, obtains the association pattern between the risk instruction and abnormal fluctuations in the power grid, and marks it as a risk warning reference.

[0039] Furthermore, the real-time monitoring and early warning module includes a real-time operational data acquisition and analysis unit and an early warning generation unit;

[0040] The real-time operation data acquisition and analysis unit collects real-time operation data of the power grid and analyzes the real-time operation data with reference to the analysis method of historical operation data to obtain the risk assessment result of the current energy storage equipment. The early warning generation unit compares and analyzes the risk assessment result of the current energy storage equipment with the corresponding historical operation dataset to match the risk early warning reference and generate corresponding prompt information.

[0041] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention, through in-depth analysis of historical power grid operation data and historical dispatch commands, combined with the health status of energy storage devices, comprehensively monitors power grid load fluctuations, frequency fluctuations, and the charging and discharging status of energy storage devices. This enables more accurate identification of abnormal power grid fluctuations and potential risks of energy storage devices. This method overcomes the limitations of traditional power grid monitoring systems that rely solely on single parameters (such as load, frequency, etc.), providing a more comprehensive risk identification capability. This invention not only analyzes power grid load and frequency fluctuations but also combines the health status of energy storage devices with power grid fluctuations to form a correlation pattern between risk commands and abnormal power grid fluctuations, thereby providing more accurate early warnings and handling suggestions. This analytical method helps predict potential risks in power grid operation, allowing for timely intervention and control measures, effectively reducing the probability of power grid failures. By assessing the real-time health status of energy storage devices and comparing it with historical operation data, this invention can promptly detect abnormal states of energy storage devices, such as overcharging or over-discharging risks, ensuring stable power grid operation. The matching analysis of real-time and historical data not only improves the accuracy of risk warnings but also enhances the system's response speed. This invention generates risk warning references related to abnormal fluctuations in the power grid and combines them with real-time monitoring data to provide intelligent early warning information for power dispatchers. This helps them identify potential power grid fault risks in advance and take corresponding measures, thereby improving the safety and stability of power grid operation. Attached Figure Description

[0042] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0043] Figure 1 This is a schematic diagram of a power grid variable monitoring system according to the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] Please see Figure 1 The present invention provides the following technical solution:

[0046] A power grid variable monitoring system includes: a data acquisition and processing module, a feature extraction and analysis module, a health assessment and risk judgment module, and a real-time monitoring and early warning module;

[0047] The data acquisition and processing module collects historical operation data and historical dispatch command records of the power grid over past time periods, analyzes the historical dispatch command records, extracts historical dispatch commands for charging and discharging of energy storage devices, summarizes and records them as a target command set; and establishes a one-to-one correspondence between the target command set and the corresponding historical operation data to construct a historical operation dataset.

[0048] The feature extraction and analysis module extracts the corresponding key trend features for each element in the historical operation dataset and constructs a key trend feature vector. Based on the key trend feature vector, load fluctuation analysis and frequency fluctuation analysis of the power grid are performed respectively to identify abnormal fluctuations.

[0049] The health assessment and risk judgment module evaluates the health status of the energy storage device corresponding to each target instruction based on the key trend feature vector; based on the assessment results of the energy storage device's health status, it determines whether the energy storage device is in a risky state of overcharging or over-discharging; if a risky state is determined, the corresponding target instruction is marked as a risk instruction, the correlation pattern between the risk instruction and abnormal fluctuations in the power grid is obtained, and it is marked as a risk warning reference.

[0050] The real-time monitoring and early warning module collects real-time operation data of the power grid, analyzes the real-time operation data with reference to the analysis method of historical operation data, and obtains the risk assessment result of the current energy storage equipment. The risk assessment result of the current energy storage equipment is compared and analyzed with the corresponding historical operation dataset to match risk warning references and generate corresponding prompt information.

[0051] The data acquisition and processing module includes a data acquisition unit and a data processing unit;

[0052] The data acquisition unit collects historical operation data and historical dispatch command records of the power grid over the past period; the data processing unit analyzes the historical dispatch command records, extracts the historical dispatch commands for charging and discharging of energy storage devices, summarizes and records them as the target command set; and maps the target command set to the corresponding historical operation data to construct a historical operation dataset.

[0053] The feature extraction and analysis module includes a feature extraction unit and an abnormal fluctuation analysis unit;

[0054] The feature extraction unit extracts the corresponding key trend features for each element in the historical operation dataset and constructs a key trend feature vector; the abnormal fluctuation analysis unit performs load fluctuation analysis and frequency fluctuation analysis of the power grid based on the key trend feature vector, thereby identifying abnormal fluctuations.

[0055] The health assessment and risk judgment module includes a health assessment unit, a risk judgment unit, and a risk instruction marking and association unit;

[0056] The health assessment unit evaluates the health status of the energy storage device corresponding to each target instruction based on the key trend feature vector; the risk judgment unit determines whether the energy storage device is in a risky state of overcharging or over-discharging based on the assessment results of the energy storage device's health status; if the risk instruction marking and association unit determines that a risky state exists, it marks the corresponding target instruction as a risk instruction, obtains the association pattern between the risk instruction and abnormal fluctuations in the power grid, and marks it as a risk warning reference.

[0057] The real-time monitoring and early warning module includes a real-time operational data acquisition and analysis unit and an early warning generation unit;

[0058] The real-time operation data acquisition and analysis unit collects real-time operation data of the power grid and analyzes the real-time operation data with reference to the analysis method of historical operation data to obtain the risk assessment result of the current energy storage equipment. The early warning generation unit compares and analyzes the risk assessment result of the current energy storage equipment with the corresponding historical operation dataset to match the risk early warning reference and generate corresponding prompt information.

[0059] A method for monitoring power grid variables includes the following steps:

[0060] Step S100. Collect historical operation data and historical dispatch command records of the power grid in the past time period, analyze the historical dispatch command records, extract the historical dispatch commands for charging and discharging of energy storage devices, summarize and record them as the target command set; match the target command set with the corresponding historical operation data one-to-one to construct the historical operation dataset;

[0061] Step S200. For each element in the historical operation dataset, extract the corresponding key trend features and construct a key trend feature vector; based on the key trend feature vector, perform load fluctuation analysis and frequency fluctuation analysis of the power grid respectively, so as to identify abnormal fluctuations;

[0062] Step S300. Based on the key trend feature vector, assess the health status of the energy storage device corresponding to each target instruction; based on the assessment results of the health status of the energy storage device, determine whether the energy storage device is in a risky state of overcharging or over-discharging; if a risky state is determined, mark the corresponding target instruction as a risk instruction, obtain the correlation pattern between the risk instruction and abnormal fluctuations in the power grid, and mark it as a risk warning reference;

[0063] Step S400. Collect real-time operation data of the power grid, analyze the real-time operation data with reference to the analysis method of historical operation data, and obtain the risk assessment result of the current energy storage device; compare and analyze the risk assessment result of the current energy storage device with the corresponding historical operation dataset, thereby matching risk warning reference and generating corresponding prompt information.

[0064] Step S100 includes:

[0065] S101. Collect historical operating data and historical dispatch command records of the power grid within a past time period. The historical operating data includes power grid load data, power grid frequency data, and energy storage system status data. The energy storage system status data includes the state of charge and charging / discharging power of the energy storage devices. From the historical dispatch command records, based on pre-set filtering keywords, select historical dispatch commands for charging and discharging of the energy storage devices and mark them as target commands. Summarize all target commands to form a target command set, represented as {I1,I2,...,In}, where I1 represents the first target command, I2 represents the second target command, and so on, In represents the nth target command, and n represents the total number of target commands within the past time period.

[0066] S102. Based on the call timestamp td of each target instruction in the target instruction set, find the historical execution data within the time period [td-T, td] and form a data pair (Ii, Di), where T represents the time step and Di represents the historical execution data corresponding to the i-th target instruction, i ranging from 1 to n; summarize all data pairs to construct a historical execution dataset S, which is represented as: S={(I1,D1),(I2,D2),...,(In,Dn)}, where D1 represents the historical execution data corresponding to the 1st target instruction, D2 represents the historical execution data corresponding to the 2nd target instruction, and so on, with Dn representing the historical execution data corresponding to the nth target instruction.

[0067] Step S200 includes:

[0068] S201. For each element in the historical operation dataset, extract the corresponding key trend features; the key trend features include load fluctuation trend features, frequency fluctuation trend features, and energy storage device status trend features; wherein the load fluctuation trend feature is the dynamic change rate of load power ΔP between adjacent time points, and the corresponding calculation formula is: ΔP=[P_load(t+Δt)-P_load(t)] / Δt, where P_load(t+Δt) represents the grid load power at time (t+Δt), P_load(t) represents the grid load power at time t, and Δt represents the time interval; similarly, the frequency fluctuation trend feature is the frequency change rate, and the frequency change rate Δfc is calculated with reference to the calculation formula of the dynamic change rate of load power ΔP; the energy storage device status trend features include the energy storage charge / discharge power ratio Pc and the state of charge change rate. S, the corresponding calculation formulas are as follows: Pc=P_charge(t) / [P_discharge(t)+∈], where P_charge(t) represents the charging power of the energy storage device at time t, P_discharge(t) represents the discharging power of the energy storage device at time t; ∈ represents the minimum value, which is greater than 0; S=[SOC(t+Δt)-SOC(t)] / Δt, where SOC(t+Δt) represents the state of charge of the energy storage device at time (t+Δt), and SOC(t) represents the state of charge of the energy storage device at time t; the extracted key trend features are averaged and normalized within the time period [td-T, td], so as to construct a key feature vector Gi for each target instruction Ii, and Gi=[ΔP_load_i,Δf_i,P_i,SCR_i];

[0069] S202. Based on the key feature vector, statistical analysis is performed on the dynamic change rate of load power ΔP_load and the frequency change rate Δf to calculate the thresholds at different confidence levels, which are respectively the upper thresholds Tp1 and Tf1, and the lower thresholds Tp2 and Tf2. For load fluctuation anomaly analysis, ΔP_load_i in the key feature vector Gi is compared with the upper threshold Tp1 and the lower threshold Tp2. If ΔP_load_i > Tp1 or ΔP_load_i < Tp2, it is preliminarily determined that there is an abnormal load fluctuation. The corresponding energy storage charge / discharge power ratio P_i and the rate of change of state of charge SCR_i are used for auxiliary judgment. When it is initially determined that there is an abnormal load fluctuation, the abnormal load fluctuation is confirmed if the following conditions are met: If ΔP_load_i > Tp1, and P_i < 1 and SCR_i < 0, it means that the energy storage device may not have responded to the increase in load in time to discharge, thus confirming the abnormal load fluctuation; if ΔP_load_i < Tp2, and P_i > 1 and SCR_i > 0, it means that the energy storage device may not have effectively utilized the excess electrical energy, thus confirming the abnormal load fluctuation.

[0070] S203. For frequency fluctuation anomaly analysis, the frequency change rate Δf_i in the key feature vector Gi is compared with the upper threshold Tf1 and the lower threshold Tf2 respectively. If Δf_i > Tf1 or Δf_i < Tf2, then a frequency fluctuation anomaly is initially present. The Pearson correlation coefficient r(Δf_i, ΔP_load_i) between the frequency change rate Δf_i and the load power dynamic change rate ΔP_load_i is calculated. If |r(Δf_i, ΔP_load_i)| > r, and a frequency fluctuation anomaly is initially present, further confirmation is made based on the direction of load and frequency changes: when ΔP_load_i > 0 and Δf_i < 0, a frequency fluctuation anomaly is confirmed; when ΔP_load_i < 0 and Δf_i > 0, a frequency fluctuation anomaly is confirmed.

[0071] In this embodiment, the conditions for confirming abnormal load fluctuations and frequency fluctuations are based on the dynamic response of the power grid system and the function of energy storage devices, as explained in detail below:

[0072] Conditions for confirming abnormal load fluctuations:

[0073] 1. Threshold determination of load power dynamic change rate:

[0074] Load fluctuation trends are measured by the dynamic rate of change of load power, ΔP_load, which reflects the instantaneous fluctuations of the grid load. If ΔP_load_i > Tp1 or ΔP_load_i < Tp2, the load fluctuation is considered to be outside the normal range, and an abnormal load fluctuation is preliminarily determined.

[0075] 2. Auxiliary judgment of energy storage devices:

[0076] When load fluctuations exceed a threshold, the status of the energy storage device is analyzed to further confirm whether there is indeed an abnormal load fluctuation. The responsiveness of the energy storage device is crucial for grid balance. If the charge-discharge power ratio P of the energy storage device is low (i.e., the energy storage device is not discharging in time to cope with the increase in load) and the rate of change of state of charge (SCR) is negative (i.e., the energy storage device is not effectively charging), then an abnormal load fluctuation can be confirmed.

[0077] Conversely, if the load fluctuation is small (ΔP_load_i < Tp2), but the charge-discharge power ratio of the energy storage device is high (i.e., the energy storage device has excess energy that has not been released in time), and the state of charge change rate is positive (i.e., the energy storage device has the ability to increase discharge), then it can also be considered that there is an abnormal load fluctuation, indicating that the energy storage device has not made full use of its stored energy to balance the load.

[0078] Conditions for confirming abnormal frequency fluctuations:

[0079] 1. Threshold determination of frequency change rate:

[0080] Frequency fluctuation trends are measured by the frequency change rate Δf. If the frequency change rate exceeds the upper threshold Tf1 or is less than the lower threshold Tf2, it is preliminarily determined that there is an abnormal frequency fluctuation, indicating that the frequency of the power grid deviates too much from the normal value.

[0081] 2. Correlation analysis between load and frequency changes:

[0082] Load and frequency are interrelated; an increase in load typically leads to a decrease in frequency, while a decrease in load typically leads to an increase in frequency. To further confirm whether there are abnormal frequency fluctuations, the Pearson correlation coefficient r(Δf,ΔPload) between the rate of change of frequency Δf and the rate of change of load power ΔPload can be calculated.

[0083] If |r(Δf,ΔPload)| is greater than a set threshold r, and the frequency decreases when the load increases (when ΔP_load_i>0 and Δf_i<0), or increases when the load decreases (ΔP_load_i<0 and Δf_i>0), then abnormal frequency fluctuations can be confirmed.

[0084] Step S300 includes:

[0085] S301. Extract the energy storage charge / discharge power ratio and state of charge change rate from the key trend feature vector, assess the health of the energy storage device corresponding to each target instruction, and obtain the comprehensive health assessment score Hi. The corresponding calculation formula is as follows:

[0086] Hi=w1·P_i+w2·SCR_i+w3·σ(P_i,SCR_i),

[0087] Where σ(P_i,SCR_i) represents the standard deviation of the charge / discharge power ratio and the rate of change of state of charge of the energy storage device, w1, w2, and w3 represent weighting coefficients, and w1+w2+w3=1; the comprehensive health assessment score Hi is compared with the threshold interval Q, and Q=[H_min,H_max], where H_min represents the minimum value of the threshold interval Q, and H_max represents the maximum value of the threshold interval Q; if the comprehensive health assessment score Hi is not within the threshold interval Q, and P_i>1 and SCR_i>0, it is judged as an overcharging risk; if the comprehensive health assessment score Hi is not within the threshold interval Q, and P_i<1 and SCR_i<0, it is judged as an over-discharging risk;

[0088] S302. Based on the judgment result, the corresponding target instruction is marked as a risk instruction; according to the identification result of abnormal fluctuation in step S200, the abnormal fluctuation timestamp tb is obtained, and the abnormal fluctuation timestamp tb is associated with the time period of the historical operation data corresponding to the risk instruction, so as to obtain the association pattern between the risk instruction and the abnormal fluctuation of the power grid, and the association pattern between the risk instruction and the abnormal fluctuation of the power grid is marked as a risk warning reference.

[0089] The correlation pattern refers to the relationship between risk commands and abnormal fluctuations in the power grid. This is achieved by associating the timestamps of abnormal fluctuations with corresponding historical operational data time periods to form a pattern. This correlation pattern helps the system identify past anomalies and trigger warnings under similar circumstances. For example, if energy storage devices pose a risk of overcharging or discharging within a specific time period, and the power grid experiences abnormal fluctuations (such as voltage or frequency fluctuations) during that period, this time period can serve as a risk warning reference. When similar abnormal power grid fluctuations occur, the system will issue corresponding warnings based on this correlation pattern.

[0090] In this embodiment, we assume that the specific data of a certain energy storage device during its charging and discharging process at a certain moment are as follows:

[0091] The energy storage device's charge / discharge power ratio Pi = 1.2, the energy storage device's state of charge change rate SCRi = 0.05, and the standard deviation calculated from historical data σ(Pi,SCRi) = 0.2; based on the above formula, the comprehensive health assessment score Hi is:

[0092] Hi = 0.4 × 1.2 + 0.4 × 0.05 + 0.2 × 0.2 = 0.48 + 0.02 + 0.04 = 0.54; the score Hi = 0.54 falls within the threshold interval Q = [0.5, 1.5], therefore it is judged that there is no risk.

[0093] Step S400 includes:

[0094] S401. Collect real-time operation data of the power grid, analyze the real-time operation data with reference to the analysis method of historical operation data, and extract the real-time key feature vector SG, where SG = [ΔPs_load, Δfs, Ps, SCRs], where ΔPs_load represents the real-time load fluctuation trend feature, Δfs represents the real-time frequency fluctuation trend feature, and Ps and SCRs represent the real-time status trend feature of the energy storage device; based on the real-time key feature vector SG, calculate the real-time health assessment score SH with reference to the calculation formula of the comprehensive health assessment score Hi; compare the real-time health assessment score SH with the threshold interval Q. If the real-time health assessment score SH is not within the threshold interval Q, and Ps > 1 and SCRs > 0, it is judged as an overcharging risk; if the comprehensive health assessment score Hi is not within the threshold interval Q, and Ps < 1 and SCRs < 0, it is judged as an over-discharging risk.

[0095] S402. Based on the risk assessment results, extract historical operational data of the same risk type from the risk instructions, calculate the similarity between the real-time operational data and the historical operational data in turn, and select the element with the highest similarity and the corresponding similarity value greater than the similarity threshold as the matching result; obtain the association pattern between the corresponding risk instructions and abnormal fluctuations in the power grid based on the matching results, and mark it as a risk warning reference; generate corresponding prompt information for relevant personnel based on the risk warning reference, and have the relevant personnel further handle the corresponding matters.

[0096] In this embodiment, based on the matching results, risk instructions and their associated patterns related to abnormal power grid fluctuations are obtained from historical operational data; the system generates warning messages related to the risks, which may include:

[0097] Risk type: "Overcharging risk"

[0098] Risk Description: "The current charging status of the grid energy storage device is abnormal, which may lead to equipment damage or safety issues. The energy storage device status Ps > 1 and SCRs > 0."

[0099] Historical case: "Based on similar historical data, overcharging issues have occurred under similar load and frequency fluctuation conditions."

[0100] The message reads: "Please check the charging status of the energy storage device immediately and adjust the charging strategy to prevent overcharging from causing equipment failure."

[0101] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0102] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method of monitoring power grid variables, characterized by: The method comprises the following steps: Step S100. Collecting historical operation data and historical scheduling instruction records of the power grid in a past time period, analyzing the historical scheduling instruction records, extracting the historical scheduling instructions of the charging and discharging of the energy storage device therefrom, and summarizing and recording as a target instruction set; and performing one-to-one correspondence between the target instruction set and the corresponding historical operation data, thereby constructing a historical operation data set; Step S200. For each element in the historical operation data set, extracting the corresponding key trend features and constructing a key trend feature vector; based on the key trend feature vector, performing load fluctuation analysis and frequency fluctuation analysis of the power grid respectively, thereby identifying abnormal fluctuations; the key trend features include load fluctuation trend features, frequency fluctuation trend features, and energy storage device state trend features; Step S300. According to the key trend feature vector, evaluating the health degree of the energy storage device corresponding to each target instruction; based on the evaluation result of the health degree of the energy storage device, judging whether the energy storage device is in a risk state of overcharging or overdischarging; if it is judged that there is a risk state, marking the corresponding target instruction as a risk instruction, obtaining the association mode of the risk instruction and the power grid abnormal fluctuation, and marking as a risk early warning reference; The step S300 comprises: S301. Extracting the energy storage charging and discharging power ratio and the state of charge change rate from the key trend feature vector, evaluating the health degree of the energy storage device corresponding to each target instruction, and thereby obtaining a comprehensive health evaluation score Hi, and the corresponding calculation formula is: Hi=w1·P_i+w2·SCR_i+w3·σ(P_i,SCR_i), wherein σ(P_i,SCR_i) represents the standard deviation of the charging and discharging power ratio and the state of charge change rate of the energy storage device, w1, w2, and w3 represent weight coefficients, and w1+w2+w3=1; comparing the comprehensive health evaluation score Hi with a threshold interval Q, and Q=[H_min,H_max], H_min represents the minimum value of the threshold interval Q, and H_max represents the maximum value of the threshold interval Q; if the comprehensive health evaluation score Hi is not within the threshold interval Q range, and P_i>1 and SCR_i>0, it is judged as an overcharging risk; if the comprehensive health evaluation score Hi is not within the threshold interval Q range, and P_i<1 and SCR_i<0, it is judged as an overdischarging risk; S302. Based on the judgment result, marking the corresponding target instruction as a risk instruction; according to the identification result of the abnormal fluctuation in step S200, obtaining an abnormal fluctuation timestamp tb, associating the abnormal fluctuation timestamp tb with the time period in which the historical operation data corresponding to the risk instruction is located, thereby obtaining the association mode of the risk instruction and the power grid abnormal fluctuation, and marking the association mode of the risk instruction and the power grid abnormal fluctuation as a risk early warning reference; Step S400. Collect real-time operation data of the power grid, analyze the real-time operation data in reference to an analysis manner of historical operation data, so as to obtain a risk judgment result of the current energy storage device; compare and analyze the risk judgment result of the current energy storage device with a corresponding historical operation data set, so as to match a risk early warning reference, and generate a corresponding prompt information.

2. A method of monitoring grid variables as claimed in claim 1, wherein: The step S100 comprises: S101. Collect historical operation data and historical scheduling instruction records in a past time period of the power grid, wherein the historical operation data comprises power grid load data, power grid frequency data and energy storage system state data; the energy storage system state data comprises energy storage device state of charge and charging and discharging power; from the historical scheduling instruction records, according to a pre-set screening keyword, historical scheduling instructions of the energy storage device charging and discharging are screened out as target instructions, all target instructions are summarized to form a target instruction set, represented as {I1, I2,..., In}, wherein I1 represents the first target instruction, I2 represents the second target instruction, and so on, and In represents the nth target instruction, and n represents a total number of target instructions in the past time period; S102. According to a call time stamp td of each target instruction in the target instruction set, historical operation data in a time period [td-T, td] is found, a data pair (Ii, Di) is formed, wherein T represents a time step, Di represents historical operation data corresponding to the ith target instruction, i takes 1 to n; all data pairs are summarized, so as to construct a historical operation data set S, and represented as: S={(I1,D1),(I2,D2),..., (In,Dn)}, wherein D1 represents historical operation data corresponding to the first target instruction, D2 represents historical operation data corresponding to the second target instruction, and so on, and Dn represents historical operation data corresponding to the nth target instruction.

3. A method of monitoring grid variables as claimed in claim 2, wherein: The step S200 comprises: S201. For each element in the historical operation data set, extract the corresponding key trend feature; wherein the load fluctuation trend feature is the dynamic change rate of the load power at adjacent time points ΔP, and the corresponding calculation formula is: ΔP=[P_load(t+Δt)-P_load(t)] / Δt, wherein P_load(t+Δt) represents the power grid load power at time (t+Δt), P_load(t) represents the power grid load power at time t, and Δt represents the time interval; similarly, the frequency fluctuation trend feature is the frequency change rate, and the frequency change rate Δfc is calculated by referring to the calculation formula of the dynamic change rate of the load power ΔP; the energy storage device state trend feature includes the energy storage charge and discharge power ratio Pc and the state of charge change rate S, and the corresponding calculation formulas are: Pc=P_charge(t) / [P_discharge(t)+Ɛ], wherein P_charge(t) represents the charging power of the energy storage device at time t, P_discharge(t) represents the discharging power of the energy storage device at time t; Ɛ represents a minimum value, which is greater than 0; S=[SOC(t+Δt)-SOC(t)] / Δt, wherein SOC(t+Δt) represents the state of charge of the energy storage device at time (t+Δt), and SOC(t) represents the state of charge of the energy storage device at time t; the extracted key trend features are averaged in the time period [td-T, td] and normalized, thereby constructing a key feature vector Gi for each target instruction Ii, and Gi=[ΔP_load_i, Δf_i, P_i, SCR_i]; S202. Based on the key feature vector, the load power dynamic change rate ΔP_load and the frequency change rate Δf are statistically analyzed, thereby calculating the threshold values under different confidence levels, and the upper threshold values Tp1 and Tf1 and the lower threshold values Tp2 and Tf2 are obtained; for load fluctuation anomaly analysis, ΔP_load_i in the key feature vector Gi is compared with the upper threshold value Tp1 and the lower threshold value Tp2, respectively, if ΔP_load_i>Tp1 or ΔP_load_i<Tp2, it is preliminarily determined that there is a load fluctuation anomaly; the corresponding energy storage charge and discharge power ratio P_i and the state of charge change rate SCR_i are introduced for auxiliary judgment, when it is preliminarily determined that there is a load fluctuation anomaly, the following conditions are met simultaneously, then the load fluctuation anomaly is confirmed: if ΔP_load_i>Tp1, and P_i<1 and SCR_i<0, the load fluctuation anomaly is confirmed; if ΔP_load_i<Tp2, and P_i>1 and SCR_i>0, the load fluctuation anomaly is confirmed; S203. For frequency fluctuation anomaly analysis, the frequency change rate Δf_i in the key feature vector Gi is compared with the upper threshold Tf1 and the lower threshold Tf2 respectively, if Δf_i>Tf1 or Δf_i 4. The method of claim 1, wherein: The step S400 comprises: S401. Collecting real-time operation data of the power grid, analyzing the real-time operation data according to the analysis mode of historical operation data, thereby extracting a real-time key feature vector SG, and SG=[ΔPs_load, Δfs, Ps, SCRs], wherein ΔPs_load represents a real-time load fluctuation trend feature, Δfs represents a real-time frequency fluctuation trend feature, Ps and SCRs represent real-time state trend features of the energy storage device; according to the real-time key feature vector SG, referring to the calculation formula of the comprehensive health assessment score Hi, calculating a real-time health assessment score SH; comparing the real-time health assessment score SH with the threshold interval Q, if the real-time health assessment score SH is not within the range of the threshold interval Q, and Ps>1 and SCRs>0, it is judged as overcharging risk; if the comprehensive health assessment score Hi is not within the range of the threshold interval Q, and Ps<1 and SCRs<0, it is judged as over-discharging risk; S402. According to the risk judgment result, extracting historical operation data of the same risk type in the risk instruction, sequentially calculating the similarity between the real-time operation data and the historical operation data, selecting the element with the maximum similarity and the corresponding similarity value greater than the similarity threshold as the matching result; obtaining the corresponding risk instruction and the correlation mode of the power grid abnormal fluctuation based on the matching result, and marking it as a risk warning reference; according to the risk warning reference, generating corresponding prompt information to relevant personnel, and further processing by relevant personnel.

5. A power grid variable monitoring system applied to the power grid variable monitoring method of any one of claims 1-4, characterized in that: The system comprises: a data acquisition and processing module, a feature extraction and analysis module, a health assessment and risk judgment module, and a real-time monitoring and early warning module; The data acquisition and processing module collects historical operation data and historical scheduling instruction records in a past time period of the power grid, analyzes the historical scheduling instruction records, extracts historical scheduling instructions of charging and discharging of the energy storage device therefrom, and summarizes and records as a target instruction set; the target instruction set and the corresponding historical operation data are one-to-one corresponding, thereby constructing a historical operation data set; The feature extraction and analysis module extracts corresponding key trend features for each element in the historical operation data set, and constructs a key trend feature vector; based on the key trend feature vector, the load fluctuation analysis and the frequency fluctuation analysis of the power grid are carried out respectively, thereby identifying abnormal fluctuations; The health assessment and risk judgment module assesses the health degree of the energy storage device corresponding to each target instruction according to the key trend feature vector; based on the assessment result of the health degree of the energy storage device, it judges whether the energy storage device is in an overcharging or over-discharging risk state; if it is judged that there is a risk state, the corresponding target instruction is marked as a risk instruction, the association mode of the risk instruction and the grid abnormal fluctuation is obtained, and it is marked as a risk warning reference; The real-time monitoring and warning module collects real-time operation data of the power grid, analyzes the real-time operation data by referring to the analysis method of historical operation data, thereby obtaining the risk judgment result of the current energy storage device; the risk judgment result of the current energy storage device is compared and analyzed with the corresponding historical operation data set, thereby matching the risk warning reference, and generating the corresponding prompt information.

6. A power grid variable monitoring system according to claim 5, characterised in that: The data acquisition and processing module includes a data acquisition unit and a data processing unit; The data acquisition unit collects historical operation data and historical scheduling instruction records in a past time period of the power grid; the data processing unit analyzes the historical scheduling instruction records, extracts the historical scheduling instructions of the energy storage device charging and discharging therefrom, and summarizes and records as a target instruction set; the target instruction set is one-to-one corresponding with the corresponding historical operation data, thereby constructing a historical operation data set.

7. A power grid variable monitoring system according to claim 5, characterized in that: The feature extraction and analysis module includes a feature extraction unit and an abnormal fluctuation analysis unit; The feature extraction unit extracts the corresponding key trend features for each element in the historical operation data set, and constructs a key trend feature vector; The abnormal fluctuation analysis unit performs load fluctuation analysis and frequency fluctuation analysis of the power grid based on the key trend feature vector, thereby identifying abnormal fluctuations.

8. A power grid variable monitoring system according to claim 5, characterized in that: The health assessment and risk judgment module includes a health assessment unit, a risk judgment unit, and a risk instruction marking and association unit; The health assessment unit assesses the health degree of the energy storage device corresponding to each target instruction according to the key trend feature vector; The risk judgment unit judges whether the energy storage device is in an overcharging or over-discharging risk state based on the assessment result of the health degree of the energy storage device; The risk instruction marking and association unit marks the corresponding target instruction as a risk instruction if it is judged that there is a risk state, obtains the association mode of the risk instruction and the grid abnormal fluctuation, and marks it as a risk warning reference.

9. A power grid variable monitoring system according to claim 5, characterized in that: The real-time monitoring and warning module includes a real-time operation data acquisition and analysis unit and a warning generation unit; The real-time operation data acquisition and analysis unit collects real-time operation data of the power grid, analyzes the real-time operation data by referring to the analysis method of historical operation data, thereby obtaining the risk judgment result of the current energy storage device; The warning generation unit compares and analyzes the risk judgment result of the current energy storage device with the corresponding historical operation data set, thereby matching the risk warning reference, and generating the corresponding prompt information.

Images