A primary frequency modulation method based on wide area information

Abstract

The present invention relates to the field of primary frequency regulation of power stations, and in particular to a primary frequency regulation method based on wide area information. The present invention dynamically adjusts the frequency of a substation power grid at the next moment based on a first predicted power, basic data and a frequency change amount, so as to adjust a target power value at the next moment by a relatively large margin, so that the adjustment speed of the power grid frequency can keep up with the change speed of the power grid load when it changes rapidly. A historical data set is classified into a plurality of power unit groups, and according to a second power average value of the power unit group at any moment in a day, 24h data of a certain date with the smallest difference from the power unit group at the same time in a day is selected from the historical data set as a reference day, so as to serve as the first predicted power of the corresponding type of power unit group at any moment. The problem that the power grid frequency regulation speed of a wide area substation may be difficult to keep up with the change speed of the power grid frequency during a period when the power grid load consumption increases or decreases rapidly is solved.

Description

Technical Field

[0001] The present invention relates to the field of primary frequency regulation of power stations, and in particular to a primary frequency regulation method based on wide area information. Background Art

[0002] By the end of 2018, the installed capacity of renewable energy in the Northwest Power Grid has reached 34.55% of the Northwest dispatching caliber installed capacity, becoming the second largest installed power source in the grid. The installed capacity and proportion are both ranked first among the six regional power grids. The proportion of conventional hydropower and thermal power units with rotational inertia has gradually decreased. Because new energy does not have the ability to respond quickly to frequency, the structural dilemma of the grid frequency control characteristics has become increasingly obvious.

[0003] At present, the main method for frequency regulation of new energy power stations is to continuously detect the grid frequency and power of each wide area substation in the new energy power station, and adjust the power to keep the frequency of the power station within the standard frequency range centered on 50Hz. However, during periods when the grid load power consumption rises or falls rapidly, the grid frequency regulation speed of the wide area substation may be difficult to keep up with the changing speed of the grid frequency, resulting in the grid frequency being difficult to meet the requirements of the grid standard frequency and difficult to comply with relevant national regulations, thereby resulting in greater cost losses. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a primary frequency modulation method based on wide-area information, which solves the problem that the grid frequency regulation speed of the wide-area substation may be difficult to keep up with the changing speed of the grid frequency during periods when the grid load power consumption rises or falls rapidly.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A primary frequency modulation method based on wide area information, the frequency modulation method comprising the following steps:

[0007] S1. Periodically collect basic data of the power grid including frequency and power through the wide area substation in the power grid;

[0008] S2. Determine whether the current frequency of the power grid in each wide area substation is within the first frequency range;

[0009] If yes, the current frequency is marked as a floating frequency and the process goes to step S4;

[0010] If not, the wide area substation whose current power grid frequency is outside the first frequency range is marked as an abnormal substation, and the process goes to step S3;

[0011] S3. Predicting the first predicted power of each wide area substation at any time according to the basic data of the abnormal wide area substation, calculating the frequency difference step of the grid frequency adjustment according to the first frequency range, and calculating the target power value of the wide area substation when adjusting at the next time;

[0012] S4. Acquire a number of continuous floating frequencies including the current moment to calculate the center frequency, and calculate a third predicted power for wide area subnet adjustment according to the change direction of the floating frequency of the center frequency.

[0013] Preferably, in step S3, the following steps are specifically included:

[0014] S31. Establish a historical data set based on basic data, and predict the first predicted power of each wide area substation at any time;

[0015] S32. Calculate the second frequency difference between the current grid frequency and the standard frequency and the frequency change in unit time according to the basic data to obtain the frequency unit change. The calculation formula of the second frequency difference is:

[0016] Δf(2)=f-f0

[0017] The formula for calculating the unit change of frequency is:

[0018]

[0019] In the above formula, Δf(2) represents the second frequency difference, f represents the current grid frequency, f0 represents the standard frequency, Δf represents the frequency unit change, and Δt represents the time taken for the frequency change to reach the second frequency difference;

[0020] S33, predicting the frequency prediction value at the next moment according to the frequency adjustment period and the frequency change;

[0021] f'=ΔfΔt'+f

[0022] In the above formula, f' represents the frequency prediction value at the next moment, and Δt' represents the frequency adjustment period, that is, the time from the completion of frequency acquisition to the completion of a frequency adjustment of the wide area substation;

[0023] S34, predicting the second predicted power at the next moment according to the frequency prediction value; the calculation formula of the second predicted power is:

[0024] P 2 =Kf'

[0025] In the above formula, P 2 represents the second predicted power, K represents the conversion coefficient, and f' represents the frequency prediction value at the next moment;

[0026] S35, calculating a predicted power difference between the second predicted power and the corresponding first predicted power;

[0027] ΔP=P 2 -P 1

[0028] In the above formula, ΔP represents the predicted power difference, P 2 represents the second predicted power at the next moment, P 1 represents the first predicted power corresponding to the next moment in the historical data set;

[0029] S36, determining whether the predicted power difference is within a power error range;

[0030] If yes, the power grid power of the wide area substation is adjusted to the second predicted power, and the process ends;

[0031] If not, proceed to the next step;

[0032] S37, correcting the second predicted power according to the power difference to obtain a target power value at the next moment; the target power value is calculated as follows:

[0033] P 0 =P 2 -DΔP

[0034] In the above formula, P 0 Indicates the target power value at the next moment, P 2 represents the second predicted power at the next moment, D represents the power adjustment coefficient, and ΔP represents the predicted power difference.

[0035] Preferably, in step S31, the following steps are specifically included:

[0036] S311, obtaining a frequency outside the standard frequency range and a corresponding time, and marking the frequency as a first abnormal frequency;

[0037] S312, according to the first frequency difference between the first abnormal frequency and the standard frequency; the calculation formula of the first frequency difference is:

[0038]

[0039] In the above formula, represents the first frequency difference, f1 represents the first abnormal frequency, and f0 represents the standard frequency, i.e. 50Hz;

[0040] S313, calculating the first power at the corresponding moment according to the first frequency difference; the calculation formula of the first power is:

[0041]

[0042] In the above formula, represents the first power at the i-th moment, K represents the conversion coefficient, represents the first frequency difference, P i represents the actual power at the i-th moment;

[0043] S314, using the first power as the historical power at the corresponding moment to establish a historical data set including each moment and the corresponding historical power;

[0044] S315, calculating the outlier degree of the historical power at each moment according to the historical data set, selecting the moment when the outlier degree is greater than the outlier degree threshold, and adjusting the historical power at the moment according to the historical power at the adjacent moments so that the outlier degree at the moment is less than the outlier degree threshold;

[0045] S316, dividing the historical data set into several power unit groups according to weather and date types, and selecting from the historical data set the date with the smallest difference from each power unit group at any time of the day as the representative day of the power unit group.

[0046] Preferably, in step S315, the following steps are specifically included:

[0047] S3151, calculating the mean value and standard deviation of the historical power in the historical data set;

[0048] S3152. Calculate the outlier degree of the historical power at each moment; the calculation formula of the outlier degree is:

[0049]

[0050] In the above formula, LQ(i) represents the outlier of the historical power at the i-th moment, P' i represents the historical power at the i-th moment in the historical data, represents the average value of the historical power in the historical data set, and σ represents the standard deviation of the historical power in the historical data set;

[0051] S3153, extract the historical power whose outlier is greater than the outlier threshold, and calculate the updated historical power at the moment according to the historical power of the two adjacent moments; the expression of the updated historical power is:

[0052] P” i =A1P' i +A2P' i-1 +A3P' i+1

[0053] In the above formula, P i represents the historical power after the update at the i-th moment in the historical data set, A1 represents the first power coefficient, P' i , P' i-1 and P'i+1 Respectively represent the historical power at the i-th moment, the historical power at the i-1-th moment, and the historical power at the i+1-th moment in the historical data set;

[0054] S3154, determining whether the outlier degree of the historical power at each moment is less than the outlier degree threshold;

[0055] If yes, proceed to step S316;

[0056] If not, return to step S3151.

[0057] Preferably, in step S316, the following steps are specifically included:

[0058] S3161. Calculate the average power of each power unit group at any time to obtain a second power average value; the calculation formula of the second power average value is:

[0059]

[0060] In the above formula, represents the second power average value of the rth power unit group at the i-th moment, P' i (r) represents the power of the r-th power unit group at the i-th moment, and the number of power values ​​in the r-th power unit group at the i-th moment is m;

[0061] S3162. Calculate the difference between any day and each power unit group based on the power at any time on any day in the historical data set and the corresponding second power average value. The calculation formula of the difference is:

[0062]

[0063] In the above formula, CY represents the difference, represents the second power average value of the rth power unit group at the i-th moment, P' i represents the power value at the i-th moment in the historical data set, and T represents the total number of moments in a day;

[0064] S3163: Set the date with the smallest difference between the historical data set and each power unit group as a representative day, and use the power at any time in the representative day as the first predicted power of each wide area substation at any time.

[0065] Preferably, in step S4, the following steps are specifically included:

[0066] S41, obtaining a number of continuous floating frequencies including the current moment;

[0067] S42, calculating the center frequency according to the floating frequency;

[0068] S43, calculating the first predicted frequency at the next moment according to the center frequency and the standard frequency range;

[0069] S44. Calculate a third predicted power for wide area subnet adjustment according to the first predicted frequency.

[0070] Preferably, in step S42, the following steps are specifically included:

[0071] S421, establishing a maximum floating interval according to the maximum and minimum values ​​of the floating frequency;

[0072] S422, dividing the maximum floating interval into a number of basic floating intervals of equal size;

[0073] S423, counting the number of floating frequencies within each basic floating interval;

[0074] S424. Obtain several basic floating intervals with the largest number of floating frequencies, and calculate the average value of their floating frequencies to obtain the center frequency.

[0075] Preferably, in step S43, the following steps are specifically included:

[0076] S431, calculating the frequency difference between the center frequency and the current frequency to obtain a first frequency difference;

[0077] S432, calculating the first predicted frequency at the next moment according to the first frequency difference; the calculation formula of the first predicted frequency at the next moment is:

[0078]

[0079] In the above formula, f(1) represents the first predicted frequency at the next moment, Δf(1) represents the first frequency difference, μ represents the frequency adjustment coefficient, Δf0 represents the allowable range of the standard frequency, and f represents the current grid frequency.

[0080] Preferably, in step S44, the following steps are specifically included:

[0081] S441, calculating the frequency difference between any moment and the previous adjacent moment to obtain a second frequency difference at any moment; the calculation formula of the second frequency difference is:

[0082] Δf(2)=f i -f i-1

[0083] In the above formula, Δf(2) is the frequency difference between any moment and the previous adjacent moment, f i and f i-1 are the frequencies of any moment and the previous adjacent moment respectively;

[0084] S442, determining whether the second frequency difference is a positive value;

[0085] If yes, mark this moment as the first moment;

[0086] If not, mark the moment as the second moment;

[0087] S443, counting the number of moments that include the current moment and are consecutive to the first moment or the second moment, that is, the first number;

[0088] S444, calculating a second predicted frequency at the next moment according to the first quantity and the first predicted frequency;

[0089] S445. Calculate a third predicted power for wide area subnet adjustment according to the second frequency; the calculation formula of the third predicted power is:

[0090] P 3 =K(f(2)-f)+P

[0091] In the above formula, P 3 represents the third predicted power, K represents the conversion coefficient, f(2) represents the second predicted frequency for wide area subnet adjustment at the next moment, and f and P are the frequency and power at the current moment, respectively. Preferably, in step S444, the calculation formula of the second predicted frequency is:

[0092]

[0093] and

[0094] Δf0(min)≤f(2)-f≤Δf0(max)

[0095] In the above formula, f(2) represents the second predicted frequency at the next moment, f represents the current grid frequency, Δf(1) represents the first frequency difference, δ represents the first quantity, ω represents the unit adjustment step, Δf0 represents the allowable range of the standard frequency, Δf0(min) represents the minimum limit value of the allowable range of the standard frequency, and Δf0(max) is the maximum limit value of the allowable range of the standard frequency.

[0096] Compared with the prior art, the present invention provides a primary frequency modulation method based on wide area information, which has the following beneficial effects:

[0097] 1. The present invention calculates the first predicted power to dynamically adjust the frequency of the substation power grid at the next moment based on the first predicted power, basic data and frequency change, so as to achieve a relatively large and accurate adjustment of the target power value at the next moment in the initial stage, so that the adjustment speed of the power grid frequency can keep up with the changing speed of the power grid load when it changes rapidly, thereby ensuring the power grid frequency in this process.

[0098] 2. The present invention analyzes the influencing factors of the first predicted power, and classifies the historical data set into several power unit groups accordingly. According to the second power average value of the power unit group at any time in a day, the 24-hour data of a certain date with the smallest difference in a day is selected from the historical data set as a reference day, so as to serve as the first predicted power of the corresponding type of power unit group at any time.

[0099] 3. The present invention calculates the outlier degree of historical data before constructing the power unit group, thereby selecting data with an outlier degree greater than an outlier degree threshold, and updating the data, thereby replacing the data to obtain a historical data set that is more in line with reality.

[0100] 4. The present invention calculates the center frequency of the floating frequency, and can significantly reduce the distance between the first predicted frequency and the center frequency at the next adjustment each time the adjustment is made, so that the number of frequency points that can be included in the interval centered on the first predicted frequency and ranging from the minimum limit value and the maximum limit value of the standard frequency allowable range is always the largest, thereby reducing the adjustment frequency of the substation. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0102] Figure 1 It is a flow chart of the primary frequency modulation method of the present invention;

[0103] Figure 2 A schematic diagram of the second predicted frequency and the first quantity of the present invention;

[0104] Figure 3 It is a schematic diagram of the wide area master station and the wide area substation of the present invention. DETAILED DESCRIPTION

[0105] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0106] Those skilled in the art can understand that all or part of the steps in the following embodiments can be completed by instructing the relevant hardware through a program, so the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0107] In order to solve the problem that the frequency regulation speed of the wide area substation may be difficult to keep up with the frequency change speed of the power grid during the period when the power grid load consumption increases or decreases rapidly, the present invention provides a primary frequency regulation method based on wide area information, such as Figure 1-Figure 3 As shown, when frequency modulation is performed, the frequency step size of the next grid adjustment is corrected based on the historical grid frequency and power data, so as to further ensure that the grid frequency can fall more accurately within the frequency standard range at the next adjustment to ensure the stability of the grid frequency. The frequency modulation method includes the following steps:

[0108] S1. The basic data of the power grid including frequency and power are collected periodically through the wide area substation in the power grid. The basic data will then be sent to the wide area master station for data processing. Finally, each wide area substation is controlled to adjust the power according to the calculated results to ensure that its power grid frequency is within the standard frequency range of the power grid. According to the provisions of the Power Supply Business Rules, the standard frequency of the power grid is 50±0.2 Hz for power grids with installed capacity of 3 million kilowatts and above, and 50±0.5 Hz for power grids with installed capacity of less than 3 million kilowatts. Under abnormal conditions of the power system, the allowable deviation of the power supply frequency is 50±1.0 Hz. The structural diagram of the wide area master station and the wide area substation is as follows Figure 3 shown.

[0109] S2. Determine whether the current frequency of the power grid in each wide area substation is within the first frequency range; the power system's primary frequency modulation response time is usually 2-20 seconds, and for systems with a capacity of 3000MW and above, a frequency deviation exceeding 50±0.2HZ is a frequency anomaly, and its duration exceeds 1 hour, which is a frequency accident; a frequency deviation exceeding 50±1HZ is an accident frequency, and its duration exceeds 15 minutes, which is a frequency accident; for systems with a capacity of less than 3000MW, a frequency deviation exceeding 50±0.5HZ is a frequency anomaly, and its duration exceeds 1 hour, which is a frequency accident; a frequency deviation exceeding 50±1HZ is an accident frequency, and its duration exceeds 15 minutes, which is a frequency accident. Therefore, in actual use, in order to ensure the time requirements of a frequency modulation and meet the requirements of not generating frequency accidents, at the same time, under the premise of the methods mentioned in steps S3 and S4, the first frequency range can be set to 50±0.4HZ, which can meet the requirements of a frequency modulation time;

[0110] If yes, the current frequency is marked as a floating frequency and the process goes to step S4;

[0111] If not, the wide area substation whose current power grid frequency is outside the first frequency range is marked as an abnormal substation, and the process goes to step S3;

[0112] S3, predicting the first predicted power of each wide area substation at any time according to the basic data of the abnormal wide area substation, and calculating the frequency difference step of the power grid frequency adjustment according to the first frequency range, and calculating the target power value of the wide area substation when adjusting at the next time; in order to further improve the accuracy of the first predicted power, the basic data of each day is classified and the power is predicted according to the influencing factors of the power grid load power consumption. In step S3, the following steps are specifically included:

[0113] S31, establish a historical data set based on the basic data, and predict the first predicted power of each wide area substation at any time; in order to make the value of the first predicted power more accurate, on the basis of studying the influencing factors of the power grid load power consumption, study the influencing factors of the first predicted power, so as to more accurately predict the first predicted power, in step S31, specifically include the following steps:

[0114] S311, obtaining a frequency outside the standard frequency range and a corresponding time, and marking the frequency as a first abnormal frequency;

[0115] S312, according to the first frequency difference between the first abnormal frequency and the standard frequency; the calculation formula of the first frequency difference is:

[0116]

[0117] In the above formula, represents the first frequency difference, f1 represents the first abnormal frequency, and f0 represents the standard frequency, i.e. 50Hz;

[0118] S313, calculating the first power at the corresponding moment according to the first frequency difference; the calculation formula of the first power is:

[0119]

[0120] In the above formula, represents the first power at the i-th moment, K represents the conversion coefficient, represents the first frequency difference, P i represents the actual power at the i-th moment;

[0121] S314, using the first power as the historical power at the corresponding moment to establish a historical data set including each moment and the corresponding historical power;

[0122] S315, calculate the outlier of the historical power at each moment according to the historical data set, select the moment when the outlier is greater than the outlier threshold, and adjust the historical power at the moment according to the historical power of the adjacent moments so that the outlier at the moment is less than the outlier threshold; in the basic data obtained, the distribution of some data may be abnormal, and the cause of the abnormality may be a sudden change in the environment or an error in data collection, so it is necessary to correct the data of this part, so the change based on the historical power is a smooth process, in step S315, specifically including the following steps:

[0123] S3151, calculating the mean value and standard deviation of the historical power in the historical data set;

[0124] S3152. Calculate the outlier degree of the historical power at each moment; the calculation formula of the outlier degree is:

[0125]

[0126] In the above formula, LQ(i) represents the outlier of the historical power at the i-th moment, P' i represents the historical power at the i-th moment in the historical data, represents the average value of the historical power in the historical data set, and σ represents the standard deviation of the historical power in the historical data set;

[0127] S3153, extract the historical power whose outlier is greater than the outlier threshold, and calculate the updated historical power at the moment according to the historical power of the two adjacent moments; the expression of the updated historical power is:

[0128] P” i =A1P' i +A2P' i-1 +A3P' i+1

[0129] In the above formula, P i represents the historical power after the update at the i-th moment in the historical data set, A1 represents the first power coefficient, P' i , P' i-1 and P' i+1 Respectively represent the historical power at the i-th moment, the historical power at the i-1-th moment, and the historical power at the i+1-th moment in the historical data set;

[0130] S3154, determining whether the outlier degree of the historical power at each moment is less than the outlier degree threshold;

[0131] If yes, proceed to step S316;

[0132] If not, return to step S3151;

[0133] S316, dividing the historical data set into several power unit groups according to weather and date types, and selecting the date with the smallest difference from each power unit group at any time of the day from the historical data set as the representative day of the power unit group; the current classification method of the power unit group is mainly to classify the date according to the holiday type, weather type and temperature of the day, the holiday types mainly include holidays, weekdays and weekends, the weather type is mainly classified into sunny, cloudy, overcast, light rain, heavy rain, and the temperature is divided into several temperature intervals, such as a certain temperature interval can be set to 20℃-25℃, so as to obtain several power unit groups, specifically, in step S316, the following steps are specifically included:

[0134] S3161. Calculate the average power of each power unit group at any time to obtain a second power average value; the calculation formula of the second power average value is:

[0135]

[0136] In the above formula, represents the second power average value of the rth power unit group at the i-th moment, P' i (r) represents the power of the r-th power unit group at the i-th moment, and the number of power values ​​in the r-th power unit group at the i-th moment is m;

[0137] S3162. Calculate the difference between any day and each power unit group based on the power at any time on any day in the historical data set and the corresponding second power average value. The calculation formula of the difference is:

[0138]

[0139] In the above formula, CY represents the difference, represents the second power average value of the rth power unit group at the i-th moment, P' i represents the power value at the i-th moment in the historical data set, and T represents the total number of moments in a day;

[0140] S3163. Set the date in the historical data set with the smallest difference from each power unit group as the representative day, and use the power at any time in the representative day as the first predicted power of each wide area substation at any time; that is, a power unit group may contain multiple dates, and each date has multiple times. At this time, the difference between the representative day obtained and the power value of the power unit group as a whole at any time is the smallest. The data of the power unit group as a whole also includes 24 hours a day and multiple times. Therefore, selecting a representative day from the historical data set is more representative than directly using the power at any time calculated by the power unit group as the first predicted power. The data of the representative day is the data that has actually been generated and is more representative.

[0141] S32. Calculate the second frequency difference between the current grid frequency and the standard frequency and the frequency change in unit time according to the basic data to obtain the frequency unit change. The calculation formula of the second frequency difference is:

[0142] Δf(2)=f-f0

[0143] The formula for calculating the unit change of frequency is:

[0144]

[0145] In the above formula, Δf(2) represents the second frequency difference, f represents the current grid frequency, f0 represents the standard frequency, Δf represents the frequency unit change, and Δt represents the time taken for the frequency change to reach the second frequency difference;

[0146] S33, predicting the frequency prediction value at the next moment according to the frequency adjustment period and the frequency change;

[0147] f'=ΔfΔt'+f

[0148] In the above formula, f' represents the frequency prediction value at the next moment, and Δt' represents the frequency adjustment period, that is, the time from the completion of frequency acquisition to the completion of a frequency adjustment of the wide area substation;

[0149] S34, predicting the second predicted power at the next moment according to the frequency prediction value; the calculation formula of the second predicted power is:

[0150] P 2 =Kf'

[0151] In the above formula, P 2represents the second predicted power, K represents the conversion coefficient, and f' represents the frequency prediction value at the next moment;

[0152] S35, calculating a predicted power difference between the second predicted power and the corresponding first predicted power;

[0153] ΔP=P 2 -P 1

[0154] In the above formula, ΔP represents the predicted power difference, P 2 represents the second predicted power at the next moment, P 1 represents the first predicted power corresponding to the next moment in the historical data set;

[0155] S36, determining whether the predicted power difference is within a power error range;

[0156] If yes, the power grid power of the wide area substation is adjusted to the second predicted power, and the process ends;

[0157] If not, proceed to the next step;

[0158] S37, correcting the second predicted power according to the power difference to obtain a target power value at the next moment; the target power value is calculated as follows:

[0159] P 0 =P 2 -DΔP

[0160] In the above formula, P 0 Indicates the target power value at the next moment, P 2 represents the second predicted power at the next moment, D represents the power adjustment coefficient, and ΔP represents the predicted power difference.

[0161] S4, obtaining a number of continuous floating frequencies including the current moment to calculate the center frequency, and calculating the third predicted power for wide area subnet adjustment according to the change direction of the center frequency and the floating frequency; taking the center frequency as the median, taking the upper and lower limits of the standard frequency range as the upper and lower limits of the center frequency, so as to include the largest number of center frequency data to the greatest extent, so as to minimize the adjustment frequency of the wide area substation. In step S4, the following steps are specifically included:

[0162] S41, obtaining a number of continuous floating frequencies including the current moment;

[0163] S42, calculating the center frequency according to the floating frequency; the center frequency is the location where the data in the floating frequency is most concentratedly distributed. In step S42, the following steps are specifically included:

[0164] S421, establishing a maximum floating interval according to the maximum and minimum values ​​of the floating frequency;

[0165] S422, dividing the maximum floating interval into a number of basic floating intervals of equal size; generally, 0.1 Hz is used as the length of a basic floating interval;

[0166] S423, counting the number of floating frequencies within each basic floating interval;

[0167] S424, obtaining a number of basic floating intervals with the largest number of floating frequencies, and calculating the average value of their floating frequencies to obtain a center frequency;

[0168] S43, calculating the first predicted frequency at the next moment according to the center frequency and the standard frequency range; because the change range of the floating frequency is small, it fluctuates around the center frequency as a whole. Therefore, in order to reduce the adjustment frequency of the wide area substation, when calculating the first predicted frequency, it is necessary to make the first predicted frequency between the center frequency and the current frequency. In step S43, the following steps are specifically included:

[0169] S431, calculating the frequency difference between the center frequency and the current frequency to obtain a first frequency difference;

[0170] S432, calculating the first predicted frequency at the next moment according to the first frequency difference; the calculation formula of the first predicted frequency at the next moment is:

[0171]

[0172] In the above formula, f(1) represents the first predicted frequency at the next moment, Δf(1) represents the first frequency difference, μ represents the frequency adjustment coefficient, Δf0 represents the allowable range of the standard frequency, and f represents the current grid frequency;

[0173] S44, calculating a third predicted power for wide area subnet adjustment according to the first predicted frequency; when the current frequency is adjusted in one direction for more than a number of times continuously, such as the frequency continues to increase, the change amplitude of the first predicted frequency is adaptively changed. In step S44, the following steps are specifically included:

[0174] S441, calculating the frequency difference between any moment and the previous adjacent moment to obtain a second frequency difference at any moment; the calculation formula of the second frequency difference is:

[0175] Δf(2)=f i -f i-1

[0176] In the above formula, Δf(2) is the frequency difference between any moment and the previous adjacent moment, f i and f i-1 are the frequencies of any moment and the previous adjacent moment respectively;

[0177] S442, determining whether the second frequency difference is a positive value;

[0178] If yes, mark this moment as the first moment;

[0179] If not, mark the moment as the second moment;

[0180] S443, counting the number of moments that include the current moment and are consecutive to the first moment or the second moment, that is, the first number;

[0181] S444. Calculate the second predicted frequency at the next moment according to the first quantity and the first predicted frequency. The calculation formula of the second predicted frequency is:

[0182]

[0183] and

[0184] Δf0(min)≤f(2)-f≤Δf0(max)

[0185] In the above formula, f(2) represents the second predicted frequency at the next moment, f represents the current grid frequency, Δf(1) represents the first frequency difference, δ represents the first quantity, ω represents the unit adjustment step, Δf0 represents the allowable range of the standard frequency, Δf0(min) represents the minimum limit value of the allowable range of the standard frequency, and Δf0(max) represents the maximum limit value of the allowable range of the standard frequency; Figure 2 As shown, when the current moment is the first moment, only the first quantity has a difference, such as Figure 2 , the first numbers of point G3, point G2 and point G1 decrease in sequence, and the second predicted frequencies corresponding to point G3, point G2 and point G1 also decrease in sequence;

[0186] S445. Calculate a third predicted power for wide area subnet adjustment according to the second frequency; the calculation formula of the third predicted power is:

[0187] P 3 =K(f(2)-f)+P

[0188] In the above formula, P 3 represents the third predicted power, K represents the conversion coefficient, f(2) represents the second predicted frequency for wide area subnet adjustment at the next moment, and f and P are the frequency and power at the current moment respectively.

[0189] The above implementation methods have been described in detail. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A primary frequency modulation method based on wide area information, characterized in that: The frequency modulation method comprises the following steps: S1. Periodically collect basic data of the power grid including frequency and power through the wide area substation in the power grid; S2. Determine whether the current frequency of the power grid in each wide area substation is within the first frequency range; If yes, the current frequency is marked as a floating frequency and the process goes to step S4; If not, the wide area substation whose current power grid frequency is outside the first frequency range is marked as an abnormal substation, and the process goes to step S3; S3. Predicting the first predicted power of each wide area substation at any time according to the basic data of the abnormal wide area substation, calculating the frequency difference step of the grid frequency adjustment according to the first frequency range, and calculating the target power value of the wide area substation when adjusting at the next time; In step S3, the following steps are specifically included: S31. Establish a historical data set based on basic data, and predict the first predicted power of each wide area substation at any time; S32. Calculate the second frequency difference between the current grid frequency and the standard frequency and the frequency change in unit time according to the basic data to obtain the frequency unit change. The calculation formula of the second frequency difference is: Δf(2)=f-f0 The formula for calculating the unit change of frequency is: In the above formula, Δf(2) represents the second frequency difference, f represents the current grid frequency, f0 represents the standard frequency, Δf represents the frequency unit change, and Δt represents the time taken for the frequency change to reach the second frequency difference; S33, predicting the frequency prediction value at the next moment according to the frequency adjustment period and the frequency change; f'=ΔfΔt'+f In the above formula, f' represents the frequency prediction value at the next moment, and Δt' represents the frequency adjustment period, that is, the time from the completion of frequency acquisition to the completion of a frequency adjustment of the wide area substation; S34, predicting the second predicted power at the next moment according to the frequency prediction value; the calculation formula of the second predicted power is: Q 2 =Kf' In the above formula, P 2 represents the second predicted power, K represents the conversion coefficient, and f' represents the frequency prediction value at the next moment; S35, calculating a predicted power difference between the second predicted power and the corresponding first predicted power; ΔP=P 2 -P 1 In the above formula, ΔP represents the predicted power difference, P 2 represents the second predicted power at the next moment, P 1 represents the first predicted power corresponding to the next moment in the historical data set; S36, determining whether the predicted power difference is within a power error range; If yes, the power grid power of the wide area substation is adjusted to the second predicted power, and the process ends; If not, proceed to the next step; S37, correcting the second predicted power according to the power difference to obtain a target power value at the next moment; the target power value is calculated as follows: P 0 =P 2 -DΔP In the above formula, P 0 Indicates the target power value at the next moment, P 2 represents the second predicted power at the next moment, D represents the power adjustment coefficient, and ΔP represents the predicted power difference; S4. Acquire a number of continuous floating frequencies including the current moment to calculate the center frequency, and calculate a third predicted power for wide area subnet adjustment according to the change direction of the floating frequency of the center frequency.

2. The primary frequency modulation method according to claim 1, characterized in that: In step S31, the following steps are specifically included: S311, obtaining a frequency outside the standard frequency range and a corresponding time, and marking the frequency as a first abnormal frequency; S312, according to the first frequency difference between the first abnormal frequency and the standard frequency; the calculation formula of the first frequency difference is: In the above formula, represents the first frequency difference, f1 represents the first abnormal frequency, and f0 represents the standard frequency, i.e. 50Hz; S313, calculating the first power at the corresponding moment according to the first frequency difference; the calculation formula of the first power is: In the above formula, represents the first power at the i-th moment, K represents the conversion coefficient, represents the first frequency difference, P i represents the actual power at the i-th moment; S314, using the first power as the historical power at the corresponding moment to establish a historical data set including each moment and the corresponding historical power; S315, calculating the outlier degree of the historical power at each moment according to the historical data set, selecting the moment when the outlier degree is greater than the outlier degree threshold, and adjusting the historical power at the moment according to the historical power at the adjacent moments so that the outlier degree at the moment is less than the outlier degree threshold; S316, dividing the historical data set into several power unit groups according to weather and date types, and selecting from the historical data set the date with the smallest difference from each power unit group at any time of the day as the representative day of the power unit group.

3. The primary frequency modulation method according to claim 2, characterized in that: In step S315, the following steps are specifically included: S3151, calculating the mean value and standard deviation of the historical power in the historical data set; S3152. Calculate the outlier degree of the historical power at each moment; the calculation formula of the outlier degree is: In the above formula, LQ(i) represents the outlier of the historical power at the i-th moment, P' i represents the historical power at the i-th moment in the historical data, represents the average value of the historical power in the historical data set, and σ represents the standard deviation of the historical power in the historical data set; S3153, extract the historical power whose outlier is greater than the outlier threshold, and calculate the updated historical power at the moment according to the historical power of the two adjacent moments; the expression of the updated historical power is: P” i =A1P' i +A2P' i-1 +A3P' i+1 In the above formula, P i represents the historical power after the update at the i-th moment in the historical data set, A1 represents the first power coefficient, P' i , P' i-1 and P' i+1 Respectively represent the historical power at the i-th moment, the historical power at the i-1-th moment, and the historical power at the i+1-th moment in the historical data set; S3154, determining whether the outlier degree of the historical power at each moment is less than the outlier degree threshold; If yes, then go to step S316; this indicates that the change of power data is relatively smooth, which is consistent with the actual change of power; If not, return to step S3151.

4. The primary frequency modulation method according to claim 2, characterized in that: In step S316, the following steps are specifically included: S3161. Calculate the average power of each power unit group at any time to obtain a second power average value; the calculation formula of the second power average value is: In the above formula, represents the second power average value of the rth power unit group at the i-th moment, P' i (r) represents the power of the r-th power unit group at the i-th moment, and the number of power values ​​in the r-th power unit group at the i-th moment is m; S3162. Calculate the difference between any day and each power unit group based on the power at any time on any day in the historical data set and the corresponding second power average value. The calculation formula of the difference is: In the above formula, CY represents the difference, represents the second power average value of the rth power unit group at the i-th moment, P' i represents the power value at the i-th moment in the historical data set, and T represents the total number of moments in a day; S3163: Set the date with the smallest difference between the historical data set and each power unit group as a representative day, and use the power at any time in the representative day as the first predicted power of each wide area substation at any time.

5. The primary frequency modulation method according to claim 1, characterized in that: In step S4, the following steps are specifically included: S41, obtaining a number of continuous floating frequencies including the current moment; S42, calculating the center frequency according to the floating frequency; S43, calculating the first predicted frequency at the next moment according to the center frequency and the standard frequency range; S44. Calculate a third predicted power for wide area subnet adjustment according to the first predicted frequency.

6. The primary frequency modulation method according to claim 5, characterized in that: In step S42, the following steps are specifically included: S421, establishing a maximum floating interval according to the maximum and minimum values ​​of the floating frequency; S422, dividing the maximum floating interval into a number of basic floating intervals of equal size; S423, counting the number of floating frequencies within each basic floating interval; S424. Obtain several basic floating intervals with the largest number of floating frequencies, and calculate the average value of their floating frequencies to obtain the center frequency.

7. The primary frequency modulation method according to claim 5, characterized in that: In step S43, the following steps are specifically included: S431, calculating the frequency difference between the center frequency and the current frequency to obtain a first frequency difference; S432, calculating the first predicted frequency at the next moment according to the first frequency difference; the calculation formula of the first predicted frequency at the next moment is: In the above formula, f(1) represents the first predicted frequency at the next moment, Δf(1) represents the first frequency difference, μ represents the frequency adjustment coefficient, Δf0 represents the allowable range of the standard frequency, and f represents the current grid frequency.

8. The primary frequency modulation method according to claim 5, characterized in that: In step S44, the following steps are specifically included: S441, calculating the frequency difference between any moment and the previous adjacent moment to obtain a second frequency difference at any moment; the calculation formula of the second frequency difference is: Δf(2)=f i -f i-1 In the above formula, Δf(2) is the frequency difference between any moment and the previous adjacent moment, f i and f i-1 are the frequencies of any moment and the previous adjacent moment respectively; S442, determining whether the second frequency difference is a positive value; If yes, mark the moment as the first moment; If not, mark the moment as the second moment; S443, counting the number of moments that include the current moment and are consecutive to the first moment or the second moment, that is, the first number; S444, calculating a second predicted frequency at the next moment according to the first quantity and the first predicted frequency; S445. Calculate a third predicted power for wide area subnet adjustment according to the second frequency; the calculation formula of the third predicted power is: P 3 =K(f(2)-f)+P In the above formula, P 3 represents the third predicted power, K represents the conversion coefficient, f(2) represents the second predicted frequency for wide area subnet adjustment at the next moment, and f and P are the frequency and power at the current moment respectively.

9. The primary frequency modulation method according to claim 8, characterized in that: In step S444, the calculation formula of the second prediction frequency is: and Δf0(min)≤f(2)-f≤Δf0(max) In the above formula, f(2) represents the second predicted frequency at the next moment, f represents the current grid frequency, Δf(1) represents the first frequency difference, δ represents the first quantity, ω represents the unit adjustment step, Δf0 represents the allowable range of the standard frequency, Δf0(min) represents the minimum limit value of the allowable range of the standard frequency, and Δf0(max) is the maximum limit value of the allowable range of the standard frequency.

Images