Virtual power plant operation platform grid-connected detection method based on power grid demand analog signal

Abstract

The invention relates to the technical field of power systems, in particular to a grid-connected detection method for a virtual power plant operation platform based on a power grid demand analog signal, which comprises the following steps of: fusing historical data, real-time data and external factors to generate a high-precision analog signal, and constructing a mathematical model of the power grid demand analog signal; quantizing grid-connected detection indexes of the virtual power plant operation platform; based on the construction result of the mathematical model and the quantification result of the virtual power plant operation platform grid-connected detection indexes, grid-connected simulation testing and fault detection are carried out on the virtual power plant operation platform, a comprehensive score is obtained, a collaborative optimization model of the target virtual power plant operation platform grid-connected strategy and control parameters is constructed, and the target virtual power plant operation platform grid-connected detection indexes are obtained. And continuously optimizing and improving the grid-connected detection model and optimizing a target virtual power plant control strategy. The problems that in the prior art, the grid-connected efficiency is not high, and the safe and stable operation risk of a power grid is large are effectively solved, the grid-connected performance of the virtual power plant is comprehensively improved, and safe and stable operation of the power grid is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a grid connection detection method for a virtual power plant operation platform based on grid demand simulation signals. Background Technology

[0002] Driven by energy transition and the "dual carbon" goal, virtual power plants have become a key force in integrating distributed energy resources and optimizing power system operation. Against the backdrop of the rapid development of smart grids and distributed energy, virtual power plants, as crucial carriers for integrating distributed energy resources and optimizing the flexibility and reliability of the power system, directly impact the safe and stable operation of the power grid through their grid connection performance.

[0003] However, existing grid connection testing methods have significant shortcomings: traditional testing relies on static test scenarios and limited data, failing to realistically simulate the dynamic demands of the power grid; testing indicators are singular, lacking a comprehensive assessment of response speed, regulation accuracy, stability, and safety; fault simulation is insufficient, making it difficult to quantify the fault tolerance capability of virtual power plants under abnormal scenarios. Furthermore, the lack of simulation signal generation methods based on multi-source data fusion leads to significant deviations between test results and actual operation.

[0004] The aforementioned problems lead to low grid connection efficiency and increased grid security risks, necessitating an innovative detection method that can comprehensively improve the grid connection performance of virtual power plants through high-precision demand simulation and multi-dimensional index analysis. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a grid connection detection method for a virtual power plant operation platform based on grid demand simulation signals, so as to solve the problems of low grid connection efficiency and high risk of grid safety and stability in the prior art.

[0006] According to a first aspect of the present invention, a method for grid connection detection of a virtual power plant operation platform based on grid demand simulation signals is provided, characterized in that the method includes:

[0007] By integrating historical data, real-time data, and external factors, a high-precision analog signal is generated, and a mathematical model for power grid demand analog signal is constructed.

[0008] The grid connection testing indicators of the virtual power plant operation platform are quantified; these indicators include: response speed, regulation accuracy, deviation rate, stability, and safety.

[0009] Based on the construction results of the mathematical model and the quantitative results of the grid connection detection indicators of the virtual power plant operation platform, grid connection simulation test and fault detection are carried out on the virtual power plant operation platform to obtain a comprehensive score.

[0010] By utilizing comprehensive scoring, a collaborative optimization model for the grid connection strategy and control parameters of the target virtual power plant operation platform is constructed. This model continuously optimizes and improves the grid connection detection model and the control strategy of the target virtual power plant.

[0011] Furthermore, the process of integrating historical data, real-time data, and external factors to generate a high-precision analog signal and constructing a mathematical model for the power grid demand analog signal includes:

[0012] Historical power grid demand data, real-time power grid operation data, and external factor data are collected, and the power grid demand data, real-time power grid operation data, and external factor data are normalized.

[0013] Based on the normalized result data, a mathematical model is constructed according to a preset formula;

[0014] The preset formula includes:

[0015] S sim =α·D hist +β·D real +γ·F ext (1)

[0016] Where α, β, and γ are dynamic weighting coefficients, α + β + γ = 1, D hist D is the normalized value of historical demand data. real F is the sliding window mean of the real-time data. ext It is a comprehensive score of external factors.

[0017] Furthermore, the method also includes;

[0018] Based on the results of the mathematical model construction, the weight allocation is optimized and the weight coefficients are dynamically adjusted using the particle swarm optimization algorithm with the objective function of minimizing the mean square error.

[0019]

[0020] Where MSE represents the mean square error, S true N represents the actual power grid demand signal, and N is the number of samples.

[0021] Furthermore, the method also includes:

[0022] The generalization ability of the mathematical model is evaluated by cross-validation. If the error exceeds the threshold, the data weights are adjusted or higher-order nonlinear terms are introduced using the following formula.

[0023]

[0024] Wherein, coefficient θ k φ k ψk It can be solved using the least squares method, and the order n, m, p is determined by cross-validation.

[0025] Furthermore, the grid connection detection indicators of the quantitative virtual power plant operation platform include:

[0026] Develop grid connection testing indicators for a virtual power plant operation platform, including response speed, adjustment accuracy, deviation rate, stability, and safety.

[0027] The response speed index in the grid connection detection index of the virtual power plant operation platform is quantified by the difference between the time from receiving the demand signal to the response time of the virtual power plant.

[0028] The adjustment accuracy index in the grid connection detection index of the virtual power plant operation platform is quantified by using the absolute deviation between the output power and the demand signal.

[0029] The deviation rate index in the grid connection detection indicators of the virtual power plant operation platform is quantified using the relative deviation percentage; the calculation method for the relative deviation percentage is as follows:

[0030]

[0031] Regarding the setting of the threshold η, according to power grid standards, η is usually required to be ≤5%; ΔP represents the absolute deviation between the output power and the demand signal;

[0032] The stability index in the grid connection testing indicators of the virtual power plant operation platform is quantified using the standard deviation of output power fluctuation; the calculation method for the standard deviation of output power fluctuation is as follows:

[0033]

[0034] Among them, P output Where μ is the output power and T is the average power over the time period.

[0035] The safety indicators in the grid connection detection index of the virtual power plant operation platform are quantified by utilizing the impact value of the grid connection process on the grid frequency; the calculation method for the impact value of the grid connection process on the grid frequency is as follows:

[0036] F dev =max (|f grid -f nominal |) (9)

[0037] Among them, f grid f is the frequency for virtual power plant grid connection. nominal This is the nominal frequency.

[0038] Furthermore, based on the construction results of the mathematical model and the quantitative results of the grid connection detection indicators of the virtual power plant operation platform, grid connection simulation tests and fault detection are conducted on the virtual power plant operation platform to obtain a comprehensive score, including:

[0039] The results of constructing the mathematical model are input into a preset virtual power plant operation platform to obtain the results of triggering grid connection response;

[0040] Based on the obtained results of the grid connection response, the output power, the time from receiving the demand signal to responding in the virtual power plant, the absolute deviation of the output power from the demand signal, the percentage of relative deviation, the standard deviation of the output power fluctuation, and the impact parameters of the grid connection process on the grid frequency are collected in real time, and the values ​​of each index are calculated.

[0041] Based on the collected parameters and the calculation results of each indicator value, a weighted scoring model is constructed and weights are allocated.

[0042] Furthermore, the weighted scoring model includes:

[0043]

[0044] Wherein, ω1 represents the response speed weight, reflecting the importance of response speed to the overall score; the larger the weight value, the higher the contribution of response speed to the score. ω2 represents the adjustment accuracy weight, reflecting the importance of adjustment accuracy to the overall score; the larger the weight value, the higher the contribution of adjustment accuracy to the score. ω3 represents the deviation rate weight, reflecting the importance of deviation rate to the overall score; the larger the weight value, the higher the contribution of deviation rate to the score. ω4 represents the stability weight, reflecting the importance of stability to the overall score; the larger the weight value, the higher the contribution of stability to the score. ω5 represents the safety weight, reflecting the importance of safety to the overall score; the larger the weight value, the higher the contribution of safety to the score.

[0045] Furthermore, the method of constructing a collaborative optimization model for the grid connection strategy and control parameters of the target virtual power plant operation platform using comprehensive scoring, and continuously optimizing and improving the grid connection detection model and the target virtual power plant control strategy, includes:

[0046] Based on the score status of the comprehensive evaluation, the grid connection status of the virtual power plant is divided into different levels;

[0047] Develop differentiated grid connection strategies for virtual power plants of different levels;

[0048] The classification standard for virtual power plants is as follows:

[0049]

[0050] Based on the grid connection strategy, a deep deterministic strategy gradient algorithm is used to optimize the virtual power plant control strategy.

[0051] The state space is as follows:

[0052] s t =[S sim ,P output ,f grid ,η,σ] (12).

[0053] Furthermore, the method also includes:

[0054] The motion space is:

[0055] a t =[ΔP setpoint ,K p ,K i ,K d (13)

[0056] Wherein, ΔP setpoint =S sim -P output ;K p The proportional gain (PG) represents the linear relationship between the control output and the current error, used to characterize the control adjustment magnitude corresponding to each unit power deviation. K p The larger the value of K, the faster the system responds to errors in real time; i The integral gain controls the relationship between the output and the accumulated historical error. It is used to characterize the adjustment weight of the accumulated error per unit time. By integrating the historical error, long-term deviations are gradually corrected, improving the adjustment accuracy. d The differential gain controls the linear relationship between the output and the rate of change of error. It is used to characterize the damping adjustment magnitude corresponding to the rate of change of error. By predicting the trend of error change, the control quantity is adjusted in advance to reduce overshoot and oscillation.

[0057] Furthermore, the method also includes:

[0058] The reward function is:

[0059] r t =Score-λ (14)

[0060] Wherein, λ is the energy consumption penalty coefficient. This function guides the virtual power plant to achieve synergistic optimization of high efficiency and energy saving during grid connection by balancing the comprehensive performance score and energy consumption control.

[0061] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0062] This invention constructs a dynamic demand simulation model, which can realistically simulate the dynamic demand of the power grid, solving the problem that traditional detection methods cannot realistically simulate the dynamic demand of the power grid. It designs multi-dimensional detection indicators to achieve a comprehensive evaluation of the response speed, regulation accuracy, stability, and safety of the virtual power plant. Combined with fault simulation and optimization algorithms, it improves the fault tolerance and grid connection performance of the virtual power plant under abnormal scenarios, effectively solving the problems of low grid connection efficiency and high risk of grid safety and stability in existing technologies, comprehensively improving the grid connection performance of the virtual power plant, and ensuring the safe and stable operation of the power grid.

[0063] Furthermore, based on the safety and stability of the power grid operation and the economy and flexibility of virtual power plant regulation, an automated matching model of operating parameters between the power grid and power plants is constructed to achieve the following objectives: accurately simulate the dynamic demand of the power grid by integrating multi-source data, quantify the multi-dimensional detection indicators of virtual power plant grid connection, and form a closed-loop optimization mechanism by combining comprehensive scoring and fault detection. Ultimately, this achieves automated adaptation between the safe and stable operation requirements of the power grid and the economical and flexible regulation capabilities of the virtual power plant, providing technical support for the efficient grid connection of virtual power plants that combines safety and economy.

[0064] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0065] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0066] Figure 1 This is a schematic diagram illustrating a grid connection detection method for a virtual power plant operation platform based on grid demand simulation signals, according to an exemplary embodiment. Detailed Implementation

[0067] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0068] Example 1

[0069] Please see Figure 1 , Figure 1 This is a schematic flowchart illustrating a grid connection detection method for a virtual power plant operation platform based on grid demand simulation signals, according to an exemplary embodiment. The method includes:

[0070] S1. Integrate historical data, real-time data, and external factors to generate high-precision analog signals and construct a mathematical model for power grid demand analog signals;

[0071] S2. Quantify the grid connection test indicators of the virtual power plant operation platform; the grid connection test indicators of the virtual power plant operation platform include: response speed, regulation accuracy, deviation rate, stability and safety indicators;

[0072] S3. Based on the construction results of the mathematical model and the quantitative results of the grid connection detection indicators of the virtual power plant operation platform, grid connection simulation test and fault detection are carried out on the virtual power plant operation platform to obtain a comprehensive score;

[0073] S4. Using comprehensive scoring, construct a collaborative optimization model of grid connection strategy and control parameters for the target virtual power plant operation platform, continuously optimize and improve the grid connection detection model, and optimize the target virtual power plant control strategy.

[0074] In practical implementation, this application provides a grid connection detection method for a virtual power plant operation platform based on grid demand simulation signals, which includes the following steps: fusing historical data, real-time data, and external factors to generate a high-precision simulation signal and constructing a mathematical model of the grid demand simulation signal; designing grid connection detection indicators, including response speed, regulation accuracy, deviation rate, stability, and safety; conducting grid connection simulation tests and fault detection on the virtual power plant operation platform through simulation testing, data analysis, and fault simulation; and formulating grid connection strategies and optimization algorithms based on the detection results to improve grid connection efficiency and system stability.

[0075] Furthermore, as described in S1, a high-precision analog signal is generated by integrating historical data, real-time data, and external factors, and a mathematical model for simulating power grid demand is constructed, specifically including:

[0076] Step 1: Data acquisition and preprocessing.

[0077] Collect historical power grid demand data D hist (Including grid load curves, renewable energy output records, market transaction data, etc.), real-time grid operation data D real (Current frequency, voltage, and power balance status of the power grid), external factor data F ext (Meteorological data (temperature, light intensity), holiday markers, electricity pricing policies, etc.). The data is normalized to eliminate dimensional differences.

[0078] Step 2: Mathematical normalization

[0079] For heterogeneous data, a minimum-maximum normalization method is used to eliminate dimensional differences:

[0080]

[0081] Where, x max x min The distribution consists of the maximum and minimum values ​​of the data characteristics.

[0082] Step 3: Mathematical Model Construction

[0083] Analog signal S sim The formula for generating it is:

[0084] S sim =α·D hist +β·D real +γ·F ext (2)

[0085] α+β+γ=1 (3)

[0086] Where α, β, and γ are dynamic weighting coefficients, and D hist D is the normalized value of historical demand data. real F is the sliding window mean of the real-time data. ext A comprehensive score for external factors (such as meteorological impact factors quantified from 0 to 1).

[0087] Step 4: Optimize weight allocation using Particle Swarm Optimization (PSO) algorithm

[0088] The weighting coefficients are dynamically adjusted with the goal of minimizing the mean squared error (MSE). The goal is to minimize the MSE between the analog signal and the actual requirement.

[0089]

[0090] Among them, S true N represents the actual power grid demand signal, and N is the number of samples.

[0091] Step 5: Signal Verification and Iterative Optimization

[0092] The model's generalization ability is evaluated through cross-validation. If the error exceeds a threshold ∈ , the data weights are adjusted or higher-order nonlinear terms (such as polynomial fitting) are introduced.

[0093]

[0094] Wherein, coefficient θ k φ k ψ k It can be solved using the least squares method, and the order n, m, p is determined by cross-validation.

[0095] Further, as described in step S2, the grid connection testing indicators of the virtual power plant operation platform are quantified; the grid connection testing indicators of the virtual power plant operation platform include: response speed, adjustment accuracy, deviation rate, stability and safety indicators, including:

[0096] Step 1: Definition and Quantification Methods of Core Detection Indicators

[0097] (1) Response speed. The time delay from when the virtual power plant receives the demand signal to when it begins to adjust its output. It reflects the virtual power plant's real-time response capability to grid commands and directly affects the grid frequency regulation effect.

[0098] The virtual power plant's response time t from receiving a demand signal resp The calculation formula is:

[0099] t resp =t start -t receive (6)

[0100] Among them, t receive t is the signal reception time. start This is the response start time.

[0101] (2) Regulation accuracy. This measures the accuracy with which the virtual power plant executes grid commands. Excessive deviation can lead to power imbalance.

[0102] The absolute deviation ΔP between the output power and the demand signal:

[0103] ΔP=|P output -S sim | (7)

[0104] Among them, P output For output power, t start This is a demand signal.

[0105] (3) Deviation rate. The relative deviation percentage, used to standardize the evaluation of the regulation performance of virtual power plants of different sizes.

[0106] Relative deviation percentage η

[0107]

[0108] Threshold setting: According to power grid standards, η is usually required to be ≤5%.

[0109] (4) Stability. The standard deviation of the output power fluctuation reflects the stability of the virtual power plant during long-term operation.

[0110] Standard deviation of output power fluctuation σ

[0111]

[0112] Where μ is the average power during the T time period.

[0113]

[0114] (5) Stability. The maximum deviation from the grid frequency during grid connection reflects the impact of the virtual power plant on grid security.

[0115] The impact of grid connection process on grid frequency F dev :

[0116] F dev =max (|f grid -f nominal |) (11)

[0117] Among them, f grid f is the frequency for virtual power plant grid connection. nominal The nominal frequency (e.g., 50Hz).

[0118] Furthermore, as described in step S3, based on the construction results of the mathematical model and the quantitative results of the grid connection detection indicators of the virtual power plant operation platform, grid connection simulation tests and fault detection are conducted on the virtual power plant operation platform to obtain a comprehensive score, including:

[0119] Step 1: Simulate the test process

[0120] (1) Signal Input. The generated S... sim Input the virtual power plant operation platform to trigger its grid connection response.

[0121] (2) Data recording. Real-time acquisition of P output t resp f grid Parameters such as these.

[0122] (3) Calculation of indicators. Calculate the values ​​of each indicator according to formulas (5) to (10).

[0123] Step 2: Comprehensive Scoring Model

[0124] To quantify overall performance, a weighted scoring model is constructed:

[0125]

[0126] Wherein, ω1 represents the response speed weight, reflecting the importance of response speed to the overall score; the larger the weight value, the higher the contribution of response speed to the score. ω2 represents the adjustment accuracy weight, reflecting the importance of adjustment accuracy to the overall score; the larger the weight value, the higher the contribution of adjustment accuracy to the score. ω3 represents the deviation rate weight, reflecting the importance of deviation rate to the overall score; the larger the weight value, the higher the contribution of deviation rate to the score. ω4 represents the stability weight, reflecting the importance of stability to the overall score; the larger the weight value, the higher the contribution of stability to the score. ω5 represents the safety weight, reflecting the importance of safety to the overall score; the larger the weight value, the higher the contribution of safety to the score.

[0127] Weighting: Set weight coefficients ω1 to ω5 according to the needs of power grid operators, and ensure that the importance of different indicators can be flexibly adjusted through weights, while avoiding scores from exceeding a reasonable range due to weight superposition.

[0128] The weighting coefficients must meet the following requirements:

[0129] ∑ω i =1 (13) Suppose the test results of a virtual power plant are as follows:

[0130]

[0131] The weights are allocated as follows:

[0132]

[0133] The overall score is:

[0134]

[0135] Step 3: Fault Simulation Scenario

[0136] (1) Voltage drop: Simulate a 20% drop in grid voltage for 0.1 seconds and check whether the virtual power plant triggers low voltage protection or power compensation.

[0137] (2) Frequency fluctuation: Set the frequency to fluctuate randomly between 49.5 and 50.5 Hz to evaluate the frequency tracking capability of the virtual power plant.

[0138] (3) Communication interruption: Simulate communication delay or packet loss to test the redundancy mechanism of the control system.

[0139] Furthermore, as described in step S4, a collaborative optimization model of the grid connection strategy and control parameters of the target virtual power plant operation platform is constructed using comprehensive scoring. This involves continuously optimizing and improving the grid connection detection model and optimizing the target virtual power plant control strategy, including:

[0140] Step 1: Tiered Grid Connection Strategy

[0141] Strategy Development Based on Comprehensive Scores: The comprehensive score is a key indicator for evaluating the grid connection performance of virtual power plants, integrating information from multiple dimensions such as response speed, regulation accuracy, deviation rate, stability, and safety. Based on the comprehensive score, virtual power plants are classified into different levels of grid connection status, and differentiated grid connection strategies are developed for virtual power plants of different levels. The classification criteria for virtual power plants are as follows:

[0142]

[0143] Virtual power plants with excellent ratings will be given priority for grid connection, with more connection time and generation quotas to fully leverage their high efficiency and stability, providing high-quality power support to the grid. Virtual power plants with good ratings will be connected to the grid normally, but their operating status needs to be continuously monitored to ensure stable performance. Virtual power plants with qualified ratings need to undergo targeted optimization and improvement before grid connection, such as adjusting equipment parameters and optimizing control algorithms, and will be connected to the grid only after a re-inspection shows a rating of good or above. Virtual power plants with unqualified ratings will be prohibited from grid connection, and their problems will be thoroughly analyzed and comprehensively rectified.

[0144] Step 2: Reinforcement Learning Optimization

[0145] The Deep Deterministic Policy Gradient (DDPG) algorithm is used to optimize the control strategy of a virtual power plant. DDPG is a reinforcement learning algorithm based on deep neural networks, suitable for optimization problems with continuous action spaces. In the virtual power plant scenario, the action space consists of the control parameters of the virtual power plant, such as adjustments to power generation and the charging and discharging strategies of the energy storage system; the state space includes real-time grid operation data (such as frequency, voltage, and power balance), the virtual power plant's own operating state (such as the output of generating equipment and the power of the energy storage system), and external environmental factors (such as meteorological data and electricity pricing policies). The core idea of ​​the algorithm is to continuously learn and optimize the control strategy through the interaction between the agent and the environment to maximize long-term cumulative rewards.

[0146] State space:

[0147] s t =[S sim ,P output ,f grid ,η,σ] (18)

[0148] By dynamically adjusting the setpoint, the virtual power plant can quickly track changes in grid demand, ensuring power balance. Specifically, ΔP setpoint The setpoint adjustment amount, i.e., the target adjustment amount of the virtual power plant's output power, is used to characterize the target value to which the current output power needs to be adjusted. sim For grid demand signals, P output This represents the current output power of the virtual power plant.

[0149] Action space:

[0150] a t =[ΔP setpoint ,K p ,K i ,K d (19)

[0151] ΔP setpoint =S sim -P output(20)

[0152] K p The proportional gain (PG) represents the linear relationship between the control output and the current error, used to characterize the control adjustment magnitude corresponding to each unit power deviation. K p The larger the value of K, the faster the system responds to errors in real time. i K is the integral gain, which controls the relationship between the output and the accumulated historical error. It is used to characterize the adjustment weight of the accumulated error per unit time. By integrating the historical error, long-term deviations are gradually corrected, and the adjustment accuracy is improved. d The differential gain controls the linear relationship between the output and the rate of change of error. It is used to characterize the damping adjustment magnitude corresponding to the rate of change of error. By predicting the trend of error change, the control quantity is adjusted in advance to reduce overshoot and oscillation.

[0153] Reward function:

[0154] r t =Score-λ (21)

[0155] This function guides the virtual power plant to achieve synergistic optimization of high efficiency and energy saving during grid connection by balancing comprehensive performance scoring and energy consumption control. Here, λ is the energy consumption penalty coefficient, used to balance the trade-off between performance improvement and energy cost.

[0156] The design of the reward function is crucial, as it directly guides the optimization direction of the virtual power plant. For example, the overall score can be used as the primary basis for rewards, while also considering factors such as grid stability and generation costs. A positive reward is given when the virtual power plant's control strategy improves the overall score, enhances grid stability, and reduces generation costs; conversely, a negative reward is given. Through continuous training, the DDPG algorithm can learn the optimal control strategy, enabling the virtual power plant to achieve efficient and stable operation under different grid conditions and external environments, thus improving overall grid connection performance.

[0157] In practical implementation, this model achieves precise hierarchical management of virtual power plant grid connection through multi-dimensional comprehensive scoring, significantly improving the utilization rate of high-quality resources. Simultaneously, through parameter adaptive optimization and a hierarchical control architecture, it effectively shortens the response time of virtual power plants and improves the accuracy of dynamically tracking grid demand. The risk constraint mechanism significantly reduces the probability of abnormal grid frequency fluctuations, and through fault scenario pre-adaptive design, it significantly enhances the system's fault recovery capability and improves grid operation safety. Furthermore, the data-driven closed-loop optimization system can accelerate the adaptation efficiency of new energy sources, improve overall grid connection efficiency, and provide core technical support for the safe and efficient access of large-scale virtual power plants.

[0158] The model's "scoring-strategy-parameter" closed-loop system breaks through the limitations of traditional static detection and extensive grid connection modes, achieving global optimization of the virtual power plant's response capability, regulation accuracy, and stability. Through the combination of intelligent algorithms and predictive control, it exhibits dynamic adaptability in power grid operation scenarios, forming intelligent management of the entire process from assessment and decision-making to control. This systematically reduces grid connection risks while maximizing the value of the virtual power plant in optimizing the flexibility and reliability of the power system.

[0159] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0160] It should be noted that in the description of this invention, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this invention, unless otherwise stated, "a plurality of" means at least two.

[0161] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0162] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0163] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0164] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0165] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0166] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0167] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A grid connection detection method for a virtual power plant operation platform based on grid demand simulation signals, characterized in that, The method includes: By integrating historical data, real-time data, and external factors, a high-precision analog signal is generated, and a mathematical model for power grid demand analog signal is constructed. Quantitative analysis of grid connection testing indicators for virtual power plant operation platforms; The grid connection testing indicators of the virtual power plant operation platform include: response speed, regulation accuracy, deviation rate, stability, and safety indicators; Based on the construction results of the mathematical model and the quantitative results of the grid connection detection indicators of the virtual power plant operation platform, grid connection simulation test and fault detection are carried out on the virtual power plant operation platform to obtain a comprehensive score. By utilizing comprehensive scoring, a collaborative optimization model for the grid connection strategy and control parameters of the target virtual power plant operation platform is constructed. This model continuously optimizes and improves the grid connection detection model and the control strategy of the target virtual power plant.

2. The method according to claim 1, characterized in that, The process of integrating historical data, real-time data, and external factors to generate a high-precision analog signal and constructing a mathematical model for power grid demand simulation signals includes: Historical power grid demand data, real-time power grid operation data, and external factor data are collected, and the power grid demand data, real-time power grid operation data, and external factor data are normalized. Based on the normalized result data, a mathematical model is constructed according to a preset formula; The preset formula includes: S sim =α·D hist +β·D real +γ·F ext (1) Where α, β, and γ are dynamic weighting coefficients, α + β + γ = 1, D hist d is the normalized value of historical demand data. real F is the sliding window mean of the real-time data. ext It is a comprehensive score of external factors.

3. The method according to claim 2, characterized in that, The method further includes; Based on the results of the mathematical model construction, the weight allocation is optimized and the weight coefficients are dynamically adjusted using the particle swarm optimization algorithm with the objective function of minimizing the mean square error. Where MSE represents the mean square error, S true N represents the actual power grid demand signal, and N is the number of samples.

4. The method according to claim 2, characterized in that, The method further includes: The generalization ability of the mathematical model is evaluated by cross-validation. If the error exceeds the threshold, the data weights are adjusted or higher-order nonlinear terms are introduced using the following formula. Wherein, coefficient θ k φ k ψ k It can be solved using the least squares method, and the order n, m, p is determined by cross-validation.

5. The method according to claim 1, characterized in that, The grid connection detection indicators of the quantitative virtual power plant operation platform include: Develop grid connection testing indicators for a virtual power plant operation platform, including response speed, adjustment accuracy, deviation rate, stability, and safety. The response speed index in the grid connection detection index of the virtual power plant operation platform is quantified by the difference between the time from receiving the demand signal to the response time of the virtual power plant. The adjustment accuracy index in the grid connection detection index of the virtual power plant operation platform is quantified by using the absolute deviation between the output power and the demand signal. The deviation rate index in the grid connection detection indicators of the virtual power plant operation platform is quantified using the relative deviation percentage; the calculation method for the relative deviation percentage is as follows: Regarding the setting of the threshold η, according to power grid standards, η is usually required to be ≤5%; ΔP represents the absolute deviation between the output power and the demand signal; The stability index in the grid connection testing indicators of the virtual power plant operation platform is quantified using the standard deviation of output power fluctuation; the calculation method for the standard deviation of output power fluctuation is as follows: Among them, P output Where μ is the output power, and T is the average power over the time period. The safety indicators in the grid connection detection index of the virtual power plant operation platform are quantified by utilizing the impact value of the grid connection process on the grid frequency; the calculation method for the impact value of the grid connection process on the grid frequency is as follows: F dev =max (|f grid -f nominal |) (9) where, f grid f is the frequency for virtual power plant grid connection. nominal This is the nominal frequency.

6. The method according to claim 1, characterized in that, Based on the construction results of the mathematical model and the quantitative results of the grid connection detection indicators of the virtual power plant operation platform, grid connection simulation tests and fault detection are conducted on the virtual power plant operation platform to obtain a comprehensive score, including: The results of constructing the mathematical model are input into a preset virtual power plant operation platform to obtain the results of triggering grid connection response; Based on the obtained results of the grid connection response, the output power, the time from receiving the demand signal to responding in the virtual power plant, the absolute deviation of the output power from the demand signal, the percentage of relative deviation, the standard deviation of the output power fluctuation, and the impact parameters of the grid connection process on the grid frequency are collected in real time, and the values ​​of each index are calculated. Based on the collected parameters and the calculation results of each indicator value, a weighted scoring model is constructed and weights are allocated.

7. The method according to claim 6, characterized in that, The weighted scoring model includes: Wherein, ω1 represents the response speed weight, reflecting the importance of response speed to the overall score; the larger the weight value, the higher the contribution of response speed to the score. ω2 represents the adjustment accuracy weight, reflecting the importance of adjustment accuracy to the overall score; the larger the weight value, the higher the contribution of adjustment accuracy to the score. ω3 represents the deviation rate weight, reflecting the importance of deviation rate to the overall score; the larger the weight value, the higher the contribution of deviation rate to the score. ω4 represents the stability weight, reflecting the importance of stability to the overall score; the larger the weight value, the higher the contribution of stability to the score. ω5 represents the safety weight, reflecting the importance of safety to the overall score; the larger the weight value, the higher the contribution of safety to the score.

8. The method according to claim 1, characterized in that, The method of using comprehensive scoring to construct a collaborative optimization model for the grid connection strategy and control parameters of the target virtual power plant operation platform, and continuously optimizing and improving the grid connection detection model and the target virtual power plant control strategy, includes: Based on the score status of the comprehensive evaluation, the grid connection status of the virtual power plant is divided into different levels; Develop differentiated grid connection strategies for virtual power plants of different levels; The classification standard for virtual power plants is as follows: Based on the grid connection strategy, a deep deterministic strategy gradient algorithm is used to optimize the virtual power plant control strategy. The state space is as follows: s t =[S sim ,P output ,f grid ,η,σ] (12).

9. The method according to claim 8, characterized in that, The method further includes: The motion space is: a t =[ΔP setpoint ,K p ,K i ,K d ] (13) Wherein, ΔP setpoint =S sim -P output ;K p The proportional gain (PG) represents the linear relationship between the control output and the current error, used to characterize the control adjustment magnitude corresponding to each unit power deviation. K p The larger the value of K, the faster the system responds to errors in real time; i The integral gain controls the relationship between the output and the accumulated historical error. It is used to characterize the adjustment weight of the accumulated error per unit time. By integrating the historical error, long-term deviations are gradually corrected, improving the adjustment accuracy. d The differential gain controls the linear relationship between the output and the rate of change of error. It is used to characterize the damping adjustment magnitude corresponding to the rate of change of error. By predicting the trend of error change, the control quantity is adjusted in advance to reduce overshoot and oscillation.

10. The method according to claim 8, characterized in that, The method further includes: The reward function is: r t =Score-λ (14) Wherein, λ is the energy consumption penalty coefficient. This function guides the virtual power plant to achieve synergistic optimization of high efficiency and energy saving during grid connection by balancing the comprehensive performance score and energy consumption control.

Images