A method for determining the inertia time constant of new energy based on delay compensation

Abstract

The present invention belongs to the technical field of inertia response of new energy, and relates to a method for determining the inertia time constant of new energy based on delay compensation. First, based on the frequency change rate and frequency offset that may exist in the power grid simulator generation system, the time nodes for the full dynamic process analysis are reasonably selected according to the key evaluation indicators of inertia to improve the analysis accuracy; secondly, the active power-frequency delay link in the inertia response process is calculated and compensated to eliminate the evaluation error caused by the delay; finally, an inertia continuous evaluation algorithm based on the differential synchronous machine swing equation is designed to achieve accurate evaluation of the inertia of new energy. This method for determining the inertia time constant of new energy based on delay compensation selects the test frequency through the maximum frequency deviation and RoCoF protection limit, and is universal. Compared with the single test RoCoF of the existing inertia response evaluation, the RoCoF considered is more comprehensive.

Description

Technical Field

[0001] The present invention belongs to the technical field of new energy inertia response, and in particular to a method for determining a new energy inertia time constant based on delay compensation. Background Art

[0002] The large-scale access of new energy sources such as wind power and photovoltaic power has become an inevitable trend in the development of future power systems, but the large-scale access of new energy sources with weak / no inertia support capabilities has brought severe challenges to the safe operation of power systems. When a large disturbance occurs, the system's ability to suppress the rate of change of frequency (RoCoF) becomes weaker, which easily triggers low-frequency load reduction actions, resulting in an increased risk of major power outages. In order to solve the problem of low system inertia, the European Union, the United Kingdom, and Australia have all put forward new requirements for the inertia response capabilities of new energy in their grid connection regulations. China has included rotational inertia in power auxiliary services in 2021, which shows that the market prospects for inertia response of new energy sources are broad in the future.

[0003] In this context, the test and evaluation of inertia response becomes increasingly important. China State Grid uses a device to simulate the grid frequency with a ramp rate of ±0.5Hz / s, and then conducts an inertia response test on the new energy virtual synchronous machine, and requires the response time to be no more than 500ms; Different from using a device to simulate the grid side frequency, the new energy inertia response test proposed by the UK directly injects virtual frequency signals with different ramp rates (±1Hz / s and ±2Hz / s) into the virtual synchronous controller, and realizes the inertia response test by analyzing the output power change. Although the above method proposes the basic idea of ​​inertia response test, the RoCoF of simulating or injecting frequency signals is relatively single, and the inertia response capability of new energy under different RoCoF is not fully tested, and the protection limit requirements of RoCoF (such as 0.2Hz / s in the UK, 0.4Hz / s in Ireland and Latvia, and 0.6Hz / s in Cyprus) are not fully considered, and there is a lack of detailed settings for test signals.

[0004] The above inertia response test focuses on the inertia power response performance and lacks quantitative evaluation of the simulated inertia value of new energy. In terms of inertia evaluation, the traditional method uses the active power and frequency data during system disturbance to evaluate the inertia offline based on the synchronous machine rotor motion equation, but the evaluation effect depends on the accurate positioning of the initial moment of the disturbance. On this basis, there is a method that uses a sliding window to process the power and frequency data to improve the positioning accuracy at the moment of disturbance, but the sliding step and window length need to be manually set based on experience, which is easy to have an adverse effect on the accuracy of the evaluation results. In addition, there is a method that uses power as input and frequency as output, establishes an autoregressive sliding average model, and realizes inertia evaluation through system identification, but the identification result depends on the modeling order of the model. In general, the above methods mainly focus on the inertia calculation at the moment of disturbance, and the inertia results based on the moment cannot accurately characterize the inertia response of the entire test process. However, the State Grid Corporation of China pointed out that the inertia response test needs to analyze the corresponding power changes during the test period (rather than the moment), and its inertia evaluation also needs to consider the entire test process, rather than using the inertia value at a certain moment. By identifying the controlled regression model and converting the result into a continuous transfer function, the continuous evaluation of inertia can be achieved through the impact response results, but there is still the problem that the model order has a great influence on the result.

[0005] In summary, there is an urgent need for an accurate inertia assessment method applicable to the entire process of new energy inertia response testing. Summary of the invention

[0006] The purpose of the present invention is to overcome the deficiencies of the prior art and to propose a method for determining the inertia time constant of a new energy source based on delay compensation.

[0007] A method for determining a new energy inertia time constant based on delay compensation, characterized in that it comprises the following steps:

[0008] Step 1: shut down other auxiliary services of the new energy power generation equipment except the inertia response function;

[0009] Step 2, according to the maximum frequency change range and frequency change rate allowed by the power system, setting the test frequency signal containing different RoCoFs;

[0010] Step 3, simulating the test frequency signal obtained in step 2 through a power grid simulator, collecting active power and frequency information at the grid connection point, judging the start and end time of the test and inertia response, and calculating the inertia response delay time;

[0011] Step 4: Based on the inertia response delay time in step 3, the active power data is delayed compensated, and the continuous identification of the equivalent inertia during the test is realized through the discretized synchronous machine swing equation. The average value of the continuous identification results under different RoCoFs is taken for evaluation.

[0012] Moreover, the method of setting the test frequency signal containing different RoCoFs in step 2 according to the maximum frequency change range and frequency change rate allowed by the power system is:

[0013] Step 2.1, based on the grid connection regulations and RoCoF relay operation, clarify the maximum frequency change range and frequency change rate allowed by the system;

[0014] Step 2.2, set the number of test groups, and based on the upper and lower limits of the maximum frequency change range determined in step 2.1, obtain the change step size of the test RoCoF, expressed as follows:

[0015]

[0016] Where: ΔRoCoF is the change step size of the test RoCoF; RoCoF max 、RoCoF min are the maximum and minimum thresholds of RoCoF anti-islanding protection constraints, respectively, and N is the number of tested RoCoF groups;

[0017] Step 2.3, based on the change step size ΔRoCoF of the test RoCoF obtained in step 2.2, determine all test RoCoF values ​​RoCoFn, and perform four groups of tests for each RoCoFn, namely, down-slope (below 50 Hz), up-slope (below 50 Hz), up-slope (above 50 Hz), and down-slope (above 50 Hz), to obtain the final test frequency signal. The method for determining RoCoFn is as follows:

[0018] RoCoF n =RoCoF min +(n-1)ΔRoCoF

[0019] Where: RoCoF n (n=1,…,N) are all tested RoCoF values.

[0020] Moreover, the step 3 simulates the test frequency signal obtained in step 2 through the power grid simulator, collects the active power and frequency information at the grid connection point, judges the start and end time of the test and the inertia response, and calculates the inertia response delay as follows:

[0021] Step 3.1, simulate the test frequency signal obtained in the second step through the power grid simulator, collect the active power and frequency information at the grid connection point, and pre-process the active power and frequency information at the grid connection point by filtering out the DC component, low-pass filtering and smoothing the data, and converting the per-unit value to eliminate the noise mixed in the measurement data, and at the same time determine the steady-state initial power of the new energy in each group of tests. The calculation formula for the steady-state initial power of the new energy in each group of tests is as follows:

[0022]

[0023] Where: P 0n,m is the initial power of steady-state new energy during the mth group climbing test of RoCoFn; T s is the power sampling time; T is the single sampling time interval; T s / T is the number of data sampling points in the sampling time period; t sn,m is the starting time of the mth group climbing test of RoCoFn; P(t) is the sampled power value at time t;

[0024] Step 3.2, based on the pre-processed data in 3.1, determine the start and end time of each test group as follows:

[0025]

[0026] Where: t bn,m ,t fn,m are the actual start and end times of the mth group test of RoCoFn respectively; q (q = 0, 1, ..., Q) is the number of sampling intervals that meet the conditions, Q is the number of consecutive times set to avoid accidental errors in the collected data and improve the anti-interference ability of the evaluation; k1 and k2 are the frequency change sensitivity coefficients respectively, the smaller k1 is and the larger k2 is, the greater t bn,m ,t fn,m The higher the sensitivity to frequency change (in this embodiment, k1 is 0.3 and k2 is 0.7); f0 is the rated frequency of the power grid;

[0027] Step 3.3, according to the active power pre-processed in step 3.1 and the steady-state initial power of the new energy during each set of inertia test, determine the start and end time of the inertia power response, the method is as follows:

[0028]

[0029] ΔP(t Bn,m )=η%ΔP refn,m

[0030] ΔP(t Fn,m )=(1-η)%ΔP refn,m

[0031] Where: ΔP refn,m For RoCoF n The incremental inertia power reference value of the mth group climbing test; H vset The inertia time constant set for the control link; t Bn,m ,t Fn,m RoCoF n The start and end times of the inertia response of the mth group of climbing tests;

[0032] Step 3.4, based on the start and end time of each test group obtained in step 3.2 and step 3.3 and the start and end time of the inertia power response, determine the inertia response delay as:

[0033]

[0034] Where: T dn,m For RoCoF n The inertia response delay of the mth group climbing test; d n,m T dn,m The multiple relationship with T is an integer.

[0035] Moreover, step 4 performs delay compensation on the active power data based on the inertia response delay time of step 3, and realizes the continuous identification of the equivalent inertia during the test process through the discretized synchronous machine swing equation. The method of taking the average value of the continuous identification results under different RoCoF for evaluation is:

[0036] Step 4.1, RoCoF obtained according to step 3 n The inertia response delay T of the mth group climbing test dn,m , delay compensation for power data is performed as follows:

[0037] P(i)=P(i+d n,m )

[0038] Among them, d n,m T dn,m It is a multiple of T and is an integer; P(i) is the active power at time t = iT

[0039] Step 4.2, discretize the synchronous machine swing equation and express it in the form of a difference equation as follows:

[0040]

[0041] Where: H v and D v are the new energy equivalent inertia time constant and virtual damping constant respectively;

[0042] Step 4.3: When performing inertia detection, turn off other auxiliary functions of the new energy source except the inertia response, so that the active power changes are all from the simulated inertia response. v is 0. On this basis, the calculation expression of the equivalent inertia time constant is obtained as follows:

[0043]

[0044] Where: H vn,m (i) RoCoF n Continuous estimate of equivalent inertia for the mth test group, i = tbn,m ,t bn,m +1,…,t fn,m ;

[0045] Step 4.4, calculate the average value of the equivalent inertia time constant obtained in step 4.3, and obtain the final inertia evaluation result as follows:

[0046]

[0047] Where: H vn,m For RoCoF n Inertia evaluation results for the mth test group, H v This is the final inertia evaluation result after the full process test.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. This new energy inertia time constant determination method based on delay compensation selects the test frequency through the maximum frequency deviation and RoCoF protection limit, which is universal. Compared with the single test RoCoF of the existing inertia response evaluation, the RoCoF considered is more comprehensive.

[0050] 2. This new energy inertia time constant determination method based on delay compensation determines the test start and end time and the inertia power response start and end time by testing the frequency signal, estimates and compensates for the inertia response delay, and improves the accuracy of the inertia response performance index and the inertia evaluation results.

[0051] 3. This new energy inertia time constant determination method based on delay compensation utilizes active power and frequency information, realizes continuous evaluation of inertia based on the discretized synchronous machine rotor motion equation, and takes its average value as the final result, avoiding the influence of different model orders on inertia evaluation in the system identification method. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 The inertia response test system diagram of the grid-connected new energy of the present invention, wherein: T1 is the grid-connected transformer of the new energy, PCC is the grid-connected point of the new energy, P pcc The new energy source at the grid-connected point generates power, S1 is the line switch of the grid simulation device, and S2 is the line switch between the grid-connected new energy source and the large grid;

[0053] Figure 2 is a schematic diagram of the inertia response performance index of the present invention, where: t a is the action time, t r is the response time, t d To adjust the time, P ref is the target power of inertia response, α is the allowable power deviation range, β is the power deviation during action, and γ is the power deviation during response;

[0054] Figure 3 This is the overall implementation flow chart of the new energy inertia time constant evaluation of the present invention.

[0055] Figure 4 It is the frequency and active power information of the whole process of the new energy inertia time constant test of the present invention.

[0056] Figure 5 This is the inertia evaluation result of the whole process of the new energy inertia time constant test of the present invention. DETAILED DESCRIPTION

[0057] In order to further understand the content, features and effects of the present invention, the following examples are listed and described in detail with reference to the accompanying drawings:

[0058] As new energy sources gradually replace traditional synchronous generators and system inertia is significantly reduced, the new inertia response of new energy sources is gradually becoming a new requirement of grid-connected regulations and a new product in the ancillary service market. The testing of its inertia response performance and the determination of the inertia time constant are of great significance for meeting the requirements of corresponding grid-connected regulations and clearing the economic benefits of ancillary service products.

[0059] The present invention realizes the identification of the equivalent inertia time constant of new energy during the test process by designing a test frequency signal and an inertia time constant identification algorithm, which helps to ensure that the inertia response of new energy meets the requirements of grid connection regulations and can also provide a basis for benefit settlement for inertia response products in the future auxiliary service market.

[0060] See also Figures 1 to 3 The innovation of the new energy inertia time constant determination method based on delay compensation (Delay Compensationbased renewable Inertia Evaluation, DCIE) lies in that it includes the following three parts: (1) test frequency signal determination; (2) inertia response delay determination; (3) equivalent inertia time constant identification. The specific steps are as follows:

[0061] Step 1: shut down other auxiliary services of the new energy power generation equipment except the inertia response function;

[0062] Step 2, according to the maximum frequency change range and frequency change rate allowed by the power system, setting the test frequency signal containing different RoCoFs;

[0063] Step 3, simulating the test frequency signal obtained in step 2 through a power grid simulator, collecting active power and frequency information at the grid connection point, judging the start and end time of the test and inertia response, and calculating the inertia response delay;

[0064] Step 4: Based on the inertia response delay result of step 3, the active power data is delayed compensated, and the continuous identification of the equivalent inertia during the test is realized through the discretized synchronous machine swing equation. The average value of the continuous identification results under different RoCoF is taken for evaluation.

[0065] The working principle of the present invention is further described below with reference to the accompanying drawings

[0066] 1. New energy inertia response test

[0067] The inertia response test simulates the frequency signal of a specific RoCoF and analyzes the active power data at the point of common coupling (PCC), such as Figure 1 As shown. Close S1, open S2, and change the actual physical frequency at the PCC point to the test frequency signal through the power grid simulator. The performance indicators of inertia response are as follows: Figure 2 The specific definitions are as follows:

[0068] Action time t a : From the time t when the set frequency changes s Start calculation, P pcc Reach β%(P ref -P0) (P0: steady-state value of output power in normal operation of new energy; Pref: target power of inertia response; β is 10).

[0069] Response time t r :P pcc Reach γ%(P ref -P0) (γ is taken as 90).

[0070] Section time t d :P pcc The time required to stabilize within the allowable deviation ±α%. (±α%: allowable deviation range, α is 2)

[0071] 2. Test frequency slope signal

[0072] The test frequency signal covers the maximum value of RoCoF of the system frequency under fault conditions, which is constrained by the RoCoF anti-islanding protection threshold of the distributed power supply. The maximum and minimum values ​​of the test signal RoCoF are set as RoCoF respectively. max 、RoCoF min According to the number of RoCoF groups N tested, the test change step size can be obtained as:

[0073]

[0074] From formula (1), we can get:

[0075] RoCoFn =RoCoF min +(n-1)ΔRoCoF (2)

[0076] Where: RoCoF n (n=1,…,N) are all test RoCoF values, n is the corresponding test group number, and N is the total number of test groups.

[0077] Secondly, considering the requirements for frequency offset when the power system is running, the maximum frequency of the test signal is set to Δf max , from which we can get the climbing test time t ramp for:

[0078]

[0079] RoCoF n Perform M groups of uphill and downhill tests, and set the start and end time of each group of tests to t sn,m ,t en,m ,but:

[0080] ten,m-tsn,m=tramp (4)

[0081] For all tests RoCoF n Four groups of up and down ramps are performed respectively, with the starting frequency being the rated frequency of the power system of 50Hz. First, a down-climbing test below 50Hz is performed, then an up-climbing test below 50Hz is performed (to restore the frequency to the rated frequency of 50Hz), then an up-climbing test above 50Hz is performed, and finally a down-climbing test above 50Hz is performed (to restore the frequency to the rated frequency of 50Hz). The ramping time of each group of tests and the time interval of each group of tests are selected as 2s. Such multiple groups of tests fully consider the various fault conditions of rising or falling when the frequency is higher or lower than 50Hz, avoiding the accidental errors of a single group of tests.

[0082] 3. Data Preprocessing

[0083] The data at the PCC point is measured at equal time intervals T, and the active power and frequency at time t = iT are taken as P(i) and f(i) respectively. Since the measured data may be mixed with noise, the data information is pre-processed by filtering out the DC component, low-pass filtering and smoothing the data, and per-unit conversion to eliminate the noise, so as to improve the accuracy of Hv evaluation.

[0084] To calculate the virtual inertia response power, we first need to know the initial power P of the new energy in steady state. 0n,m . Take t sm,n Front T s The average active power in the time period is used as RoCoF n P of the mth group of hill climbing test0n,m , and its calculation model is:

[0085]

[0086] This model further eliminates the noise of data acquisition. Then, the incremental inertia power during each test period can be obtained:

[0087] ΔP(i)=P(i)-P 0n,m (t sn,m ≤t≤t en,m ) (6)

[0088] 4. Inertia response delay compensation

[0089] The frequency change generated by the grid simulator at the PCC point cannot modulate the grid frequency defined by the input signal with a specific frequency change rate in real time due to control delay, which may cause the ramp rate to fail to reach the specific RoCoF immediately. n If the start and end time t is set sn,m ,t en,m Therefore, it is necessary to measure the frequency signal at the PCC point and calculate the actual start and end time t of each set of inertia response tests. bn,m and t fn,m The specific method is as follows:

[0090]

[0091] In the formula: q (q = 0, 1, ..., Q) is the number of sampling intervals that meet the conditions, Q is the number of consecutive settings to avoid accidental errors in the collected data and improve the anti-interference ability of the evaluation; k1 and k2 are the frequency change sensitivity coefficients, the smaller k1 is and the larger k2 is, the greater t bn,m ,t fn,m The higher the sensitivity to frequency change (in this embodiment, k1 is 0.3 and k2 is 0.7); f0 is the rated frequency of the power grid.

[0092] After satisfying formula (7), t is determined bn,m ,t fn,m , replace t in equations (8) and (9) sn,m ,t en,m , eliminating the start and end time errors caused by the control delay of the power grid simulator.

[0093] New energy does not have a real rotor, but it can achieve virtual inertia power output under virtual inertia control, expressed as:

[0094]

[0095] Where: ΔP pccis the per unit value of active power increment at PCC; H vset For the set new energy equivalent inertia time constant, we can further obtain:

[0096]

[0097] Where: ΔP refn,m For RoCoF n The incremental inertia power reference value of the mth group of climbing tests.

[0098] Frequency starts to change until the active power increment rises to η%ΔP refn,m The time is the response delay (η is 10). Then the start and end time of the inertia response is t Bn,m ,t Fn,m Don't:

[0099] ΔP(t Bn,m )=η%ΔP refn,m

[0100] ΔP(t Fn,m )=(1-η)%ΔP refn,m (10)

[0101] According to the determined t bn,m ,t fn,m , inertia response delay T dn,m The calculation method is:

[0102]

[0103] Where: d n,m T dn,m The multiple relationship with T is an integer.

[0104] 5. New energy equivalent inertia

[0105] The time domain expression of the synchronous machine swing equation is:

[0106]

[0107] Where: H v and D v are the new energy equivalent inertia time constant and virtual damping constant respectively.

[0108] Transformed into a discrete model, it can be expressed in the form of a difference equation:

[0109]

[0110] After finishing, we can get:

[0111] When performing inertia detection, turn off all auxiliary functions of the new energy source except the inertia response, so that the active power changes are all from the simulated inertia response. v is 0. On this basis, the calculation expression of the equivalent inertia time constant is obtained as follows:

[0112]

[0113] According to formula (11), the delay T is obtained dn,m , and delay compensation is performed on equation (14):

[0114]

[0115] Where: H vn,m (i) RoCoF n The continuous estimate of equivalent inertia of the mth test group (i = t bn,m ,t bn,m +1,…,t fn,m ).

[0116] Then the equivalent inertia H of this group of tests is obtained vn,m for:

[0117]

[0118] For each group H vn,m Find the average value and get the final inertia evaluation result H v for:

[0119]

[0120] Method performance verification

[0121] like Figure 1 The simulation system shown uses the above method to evaluate inertia, where photovoltaics are used as new energy. The electrical parameters of the photovoltaic grid-connected system and the grid simulator are shown in Table 1:

[0122] RoCoF of example settings max 、RoCoF min and the number of test groups N are 0.5Hz / s, 0.1Hz / s, 3, Δf max The 1Hz is the H in the photovoltaic inertia simulation control link. vset Set to 6s.

[0123] The proposed new energy inertia time constant determination method based on delay compensation (named DCIE) and the traditional method based on system model identification (named COMP) are compared. The frequency and power information of the whole process are as follows: Figure 4 As shown, the test frequency RoCoF nThe values ​​were 0.1 Hz / s, 0.3 Hz / s and 0.5 Hz / s respectively. Four ramp tests were performed in each group, e.g. Figure 4 As shown by the red solid line, the corresponding inertia response power is Figure 4 The blue dashed line shows the inertia identification result. Figure 5 As shown in the figure, the black dotted line is the actual inertia time constant of the new energy inertia simulation control (i.e. the set value H vset =6s), the dynamic results of the inertia time constant evaluated by the COMP method are generally higher than those of the H vset , while the inertia time constant dynamic evaluation result of the DCIE method is closer to H vset , with higher accuracy. Compared with the COMP method, DCIE also has faster convergence speed and higher stability;

[0124] The statistical analysis results are shown in Table 2. The final results of the inertia time constant evaluation of the DCIE and COMP methods are 5.666s and 7.322s, respectively, with errors of 5.57% and 22.03%, respectively. It can be seen that the error of the DCIE method is significantly lower than that of the comparison method.

[0125] Table 1 Parameters of photovoltaic grid-connected system and grid simulator

[0126] parameter Numeric <![CDATA[Rated capacity S of the photovoltaic system n > 3MVA <![CDATA[Photovoltaic system operating output P0]]> 0.7pu <![CDATA[DC side voltage u of the photovoltaic system dc0 > 1.2kV <![CDATA[AC-side voltage u of the photovoltaic system ac0 > 0.69kV Photovoltaic system AC side filter inductor 0.1pu Photovoltaic system AC side filter resistor 0.0015pu <![CDATA[Supercapacitor C sc > 2.46F / MW <![CDATA[Ultra-capacitor energy storage inductor L sc > <![CDATA[9.88×10 -7 H]]> <![CDATA[Supercapacitor energy storage resistor R sc > <![CDATA[3.05×10 -5 Oh<!-- 8 --> ]]> Transformer T1 rated capacity 3.6MVA Transformer T1 ratio 35 / 0.69kV Grid simulator rated capacity 5MVA

[0127] Table 2 Statistical analysis results of the whole process

[0128]

[0129] The embodiments described above are only used to illustrate the technical ideas and features of the present invention, and their purpose is to enable those skilled in the art to understand the contents of the present invention and implement them accordingly. The patent scope of the present invention cannot be limited only by this embodiment, that is, any equivalent changes or modifications made to the spirit disclosed by the present invention still fall within the patent scope of the present invention.

Claims

1. A method for determining the inertia time constant of a new energy source based on delay compensation, characterized in that: The steps include: Step 1: shut down other auxiliary services of the new energy power generation equipment except the inertia response function; Step 2, according to the maximum frequency change range and frequency change rate allowed by the power system, setting the test frequency signal containing different RoCoFs; Step 3, simulating the test frequency signal obtained in step 2 through a power grid simulator, collecting active power and frequency information at the grid connection point, judging the start and end time of the test and inertia response, and calculating the inertia response delay time; Step 4: based on the inertia response delay time in step 3, delay compensation is performed on the active power data, and the continuous identification of the equivalent inertia during the test is realized through the discretized synchronous machine swing equation. The average value of the continuous identification results under different RoCoFs is taken for evaluation; The method of setting the test frequency signal containing different RoCoFs in step 2 according to the maximum frequency change range and frequency change rate allowed by the power system is: Step 2.1, based on the grid connection regulations and RoCoF relay operation, clarify the maximum frequency change range and frequency change rate allowed by the system; Step 2.2, setting the number of test groups, and obtaining the change step length ΔRoCoF of the test RoCoF based on the upper and lower limits of the maximum frequency change range determined in step 2.1; Step 2.3, based on the change step size ΔRoCoF of the test RoCoF obtained in step 2.2, determine all test RoCoF values ​​RoCoFn, and perform four groups of tests on each RoCoFn, namely, down-climbing below 50 Hz, up-climbing below 50 Hz, up-climbing above 50 Hz, and down-climbing above 50 Hz, to obtain the final test frequency signal; The method of step 3 in which the test frequency signal obtained in step 2 is simulated by a power grid simulator, active power and frequency information at the grid connection point are collected, the start and end time of the test and the inertia response are judged, and the inertia response delay is calculated as follows: Step 3.1, simulating the test frequency signal obtained in the second step through the power grid simulator, collecting the active power and frequency information at the grid connection point, and pre-processing the active power and frequency information at the grid connection point by filtering out the DC component, low-pass filtering and smoothing the data, and converting the data into per-unit value to eliminate the noise mixed in the measurement data, and at the same time determining the steady-state initial power of the new energy in each group of tests; Step 3.2, judging the start and end time of each test group according to the active power and frequency information preprocessed in step 3.1; Step 3.3, judging the start and end time of the inertia power response according to the active power preprocessed in step 3.1 and the steady-state initial power of the new energy during each set of inertia test; Step 3.4, determining the inertia response delay time according to the start and end time of each test group obtained in step 3.2 and step 3.3 and the start and end time of the inertia power response; The step 4 performs delay compensation on the active power data based on the inertia response delay result of step 3, and realizes the continuous identification of the equivalent inertia during the test process through the discretized synchronous machine swing equation. The method of taking the average value of the continuous identification results under different RoCoF for evaluation is: Step 4.1, RoCoF obtained according to step 3 n The inertia response delay T of the mth group climbing test dn,m , delay compensation for power data is performed as follows: P(i)=P(i+d n,m ) Among them, d n,m T dn,m It is a multiple of T and is an integer; P(i) is the active power at time t = iT Step 4.2, discretize the synchronous machine swing equation and express it in the form of a difference equation as follows: Where: H v and D v are the new energy equivalent inertia time constant and virtual damping constant respectively; Step 4.3: When performing inertia detection, turn off other auxiliary functions of the new energy source except the inertia response, so that the active power changes are all from the simulated inertia response. v is 0. On this basis, the calculation expression of the equivalent inertia time constant is obtained as follows: Where: H vn,m (i) RoCoF n Continuous estimate of equivalent inertia for the mth test group, i = t bn,m ,t bn,m +1,…,t fn,m ; Step 4.4, calculate the average value of the equivalent inertia time constant obtained in step 4.3, and obtain the final inertia evaluation result as follows: Where: H vn,m For RoCoF n Inertia evaluation results for the mth test group, H v is the final inertia evaluation result after the full process test, t bn,m ,t fn,m are the actual start and end times of the mth group test of RoCoFn, and N is the number of RoCoF groups tested.

2. The method for determining the inertia time constant of a new energy source based on delay compensation according to claim 1, characterized in that: The expression for obtaining the change step length of the test RoCoF is as follows: Where: ΔRoCoF is the change step size of the test RoCoF; RoCoF max 、RoCoF min are the maximum and minimum thresholds of RoCoF anti-islanding protection constraints respectively, and N is the number of tested RoCoF groups.

3. The method for determining the inertia time constant of a new energy source based on delay compensation according to claim 1 is characterized in that: The RoCoFn is determined as follows: RoCoF n =RoCoF min +(n-1)ΔRoCoF Where: RoCoF n (n=1,…,N) are all tested RoCoF values.

4. The method for determining the inertia time constant of a new energy source based on delay compensation according to claim 1, characterized in that: The calculation formula of the steady-state initial power of the new energy in each test is as follows: Where: P 0n,m is the initial power of steady-state new energy during the mth group climbing test of RoCoFn; T s is the power sampling time; T is the single sampling time interval; T s / T is the number of data sampling points in the sampling time period; t sn,m is the starting time of the mth group climbing test of RoCoFn; P(t) is the sampled power value at time t.

5. The method for determining the inertia time constant of a new energy source based on delay compensation according to claim 1, characterized in that: The method for determining the start and end time of each test group is as follows: Where: t bn,m ,t fn,m are the actual start and end times of the mth group test of RoCoFn respectively; q, q = 0, 1, ..., Q is the number of sampling intervals that meet the conditions, Q is the number of consecutive times set to avoid accidental errors in data collection and improve the anti-interference ability of the evaluation; k1 and k2 are the frequency change sensitivity coefficients respectively, the smaller k1 is and the larger k2 is, the greater t bn,m ,t fn,m The higher the sensitivity to frequency changes; f0 is the rated frequency of the power grid.

6. The method for determining the inertia time constant of a new energy source based on delay compensation according to claim 1, characterized in that: The method for determining the start and end time of the inertia power response is as follows: ΔP(t Bn,m )=η%ΔP refn,m ΔP(t Fn,m )=(1-η)%ΔP refn,m Where: ΔP refn,m For RoCoF n The incremental inertia power reference value of the mth group climbing test; H vset The inertia time constant set for the control link; t Bn,m ,t Fn,m RoCoF n The start and end times of the inertia response of the mth group of hill climbing tests.

7. The method for determining the inertia time constant of a new energy source based on delay compensation according to claim 1, characterized in that: The inertia response delay time is determined as: Where: T dn,m For RoCoF n The inertia response delay of the mth group climbing test; d n,m T dn,m The multiple relationship with T is an integer.

Images