A fast and reliable frequency modulation method, device and medium based on energy storage

Abstract

The application discloses a kind of based on energy storage's fast reliable frequency modulation method, equipment and medium, method includes: by frequency detection device real-time acquisition power grid frequency, and real-time acquisition battery SOC value in energy storage system;According to the SOC value obtained, grading is carried out, and the battery state is divided into first number grade;Grid frequency is filtered and handled to calculate moving average frequency, and then the grid frequency is graded, and the grid frequency is divided into second number grade;Frequency difference value is calculated based on frequency moving average value;Based on SOC grading and frequency grading, frequency control is realized in combination with frequency difference value.The application realizes energy storage system in fast response power grid frequency modulation demand simultaneously, adaptive optimization charging and discharging strategy by "multi-dimension grading+steady state control", avoid overcharge and overdischarge, improve energy storage utilization and power grid stability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power batteries, in particular to a fast and reliable frequency regulation method based on energy storage, a device and a medium. BACKGROUND

[0002] With the large-scale access of new energy to the power grid, especially the fluctuation of new energy forms such as wind energy and solar energy, the frequency fluctuation of the power grid becomes more frequent and severe. In order to maintain the stability of the power grid, energy storage power stations are usually used for frequency regulation. Energy storage power stations have a fast response speed and can respond to frequency changes within 500 milliseconds to provide fast power regulation.

[0003] However, the existing frequency regulation scheme of energy storage power stations still has some deficiencies, mainly in the following aspects:

[0004] 1. Response speed and stability contradiction: The existing algorithm does not consider the SOC boundary state of the energy storage (such as forced high-power discharge when the power is low) when pursuing fast response, resulting in overcharge / overdischarge of the energy storage, affecting the service life of the power station and the stability of the power grid.

[0005] 2. Low utilization rate of energy storage: The existing frequency regulation algorithm is mostly based on a pre-set static response strategy, without real-time optimization of charging and discharging power according to the SOC, such as not actively reducing the charging power when the SOC is high, resulting in waste of energy storage resources; not charging slowly in advance when the SOC is low, missing the opportunity for frequency regulation. It may not be able to maximize the performance of the energy storage power station in some periods.

[0006] 3. Risk of power grid impact: When the SOC triggers the alarm threshold, the traditional algorithm often uses a "cliff" power switching (such as suddenly stopping discharging), resulting in secondary fluctuations in the power grid frequency.

[0007] 4. Lack of hierarchical control logic: SOC is not divided into "normal-warning-alarm" multi-level states, and it is not possible to achieve "emergency frequency regulation priority" and "daily power maintenance" coordination. The existing mode may cause the battery to be overcharged or overdischarged when the energy storage power station is frequency-regulated, affecting the service life of the energy storage battery and the effect of frequency regulation.

[0008] There are also existing schemes that use DDPG deep learning algorithms or dynamic allocation based on reinforcement learning, but they rely on complex models and have high computational costs, poor real-time performance, and rely on a large amount of training data. SUMMARY

[0009] Technical purposes: In view of the defects in the prior art, the application discloses a kind of based on energy storage's fast reliable frequency modulation method, equipment and medium, by " multi-dimension grading + steady control ", realize energy storage system in the fast response grid frequency modulation demand, while adaptive optimization charging and discharging strategy, avoid overcharge overdischarge, improve energy storage utilization and grid stability.

[0010] Technical scheme: To achieve the above technical purposes, the application adopts the following technical solutions.

[0011] A kind of based on energy storage's fast reliable frequency modulation method, method includes:

[0012] The grid frequency is acquired in real time by frequency detection device, and the SOC value of battery in energy storage system is acquired in real time;

[0013] According to the SOC value obtained, grading is carried out, and the battery state is divided into a first number of grades;

[0014] The moving average frequency is calculated by filtering and processing the grid frequency, and then the grid frequency is graded, and the grid frequency is divided into a second number of grades;

[0015] The frequency difference value is calculated based on the moving average frequency;

[0016] Based on SOC grading and frequency grading, frequency control is realized in combination with the frequency difference value.

[0017] Further, the frequency filtering process includes: according to the obtained grid frequency, the moving average frequency is calculated according to the first preset length The calculation formula of the moving average frequency

[0018] ,

[0019] Wherein, is the moving average frequency; is the real-time frequency value of the i th sampling;N represents the number of sampling points.

[0020] Further, based on SOC grading and frequency grading, frequency control is realized in combination with the frequency difference value, including:

[0021] When the battery is in normal state, the energy storage system outputs linearly according to frequency change, sets linear parameters according to frequency grading , [49.9, 50.1] is set as the first frequency region, [49.5, 49.9] and [50.1, 50.5] are set as the second frequency region, the system outputs the first current energy storage system according to the first current energy storage system when it is in the first frequency region and the first frequency region, when the moving average value of frequency is lower than 49.5 and higher than 50.5, the system outputs the maximum frequency.​

[0022] When the battery is in the charging pre-warning state or the discharging pre-warning state, the energy storage system linearly outputs power according to the frequency change, and simultaneously allows compensation of the battery power, sets a pre-warning state compensation coefficient according to the frequency grading, and calculates a second current energy storage system output power in combination with the first current energy storage system output power;

[0023] When the battery is in the charging warning state or the discharging warning state, the energy storage system only provides instantaneous adjustment when the frequency changes, and exits after the frequency is stable.

[0024] Further, the calculation formula of the first current energy storage system output power comprises:

[0025] ,

[0026] wherein, is the current energy storage system output power; is a linear parameter, which is set according to the frequency grading, is a frequency difference.

[0027] Further, the calculation formula of the second current energy storage system output power comprises:

[0028] ,

[0029] wherein, is the current time is the pre-warning state variable of the nth point in 300 time points within the previous 5 minutes, when the battery is in the charging pre-warning state, , indicates that the system is in a high pre-warning state, when the battery is in the discharging pre-warning state, , indicates that the system is in a low pre-warning state, otherwise 0, indicating that the system is not in a pre-warning state; is a pre-warning state compensation coefficient, which is set according to the frequency grading, is the current energy storage system output power, is the first current energy storage system output power.

[0030] Further, the calculation formula of the third current energy storage system output power comprises:

[0031] ,

[0032] wherein, is the current time is the moving average of the nth point in 300 time points within the previous 5 minutes, is a moving average frequency, is the current energy storage system output power.

[0033] Further, the first number is five, including a charging warning state, a charging early warning state, a normal state, a discharging warning state, and a discharging early warning state.

[0034] Further, the second number is three, including a high state, a normal state, and a low state; based on the calculated frequency moving average Classification is performed; when the frequency interval is [50.1, 50.5], the frequency is in a “high” state; when the frequency interval is [49.9, 50.1], the frequency is in a “normal” state; and when the frequency interval is [49.9, 49.5], the battery is in a “low” state.

[0035] The application further discloses a computer device, which comprises a processor and a memory, the memory stores a computer program, and the processor is used for executing the computer program to implement the above-mentioned energy storage-based fast and reliable frequency regulation method.

[0036] The application further discloses a computer storage medium, which stores a computer program, and the computer program implements the above-mentioned energy storage-based fast and reliable frequency regulation method when executed on a processor.

[0037] Beneficial effects: the application proposes a model-independent analytical control framework, realizes the collaborative optimization of SOC and frequency while ensuring real-time through regular classification and prediction correction, realizes the self-adaptive optimization of charging and discharging strategies while the energy storage system quickly responds to the frequency regulation demand of the power grid through “multi-dimensional classification + steady-state control”, avoids overcharging and overdischarging, and improves the utilization rate of energy storage and the stability of the power grid. The application realizes the collaborative optimization of fast response and safety protection by fusing SOC state prediction and frequency response characteristics. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 The method flowchart of the application. DETAILED DESCRIPTION

[0039] In order for those skilled in the art to better understand the application, the technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only some of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0040] Embodiment 1

[0041] As shown in the accompanying Figure 1As shown, the energy storage-based fast and reliable frequency modulation method of the embodiment includes the following steps:

[0042] S1, acquiring the grid frequency in real time through a frequency detection device, and acquiring the SOC value of the battery in the energy storage system in real time; wherein the grid frequency f is collected in real time according to a preset sampling frequency, such as 10Hz, and the high-frequency collection frequency can realize the high-speed response speed of the system; the SOC value of the battery in the energy storage system is acquired in real time according to a preset sampling frequency, such as 1Hz;

[0043] S2, classifying according to the acquired SOC value, and dividing the battery state into a first number of levels; in the embodiment, the first number is five, including a charging warning state, a charging early warning state, a normal state, a discharging warning state, and a discharging early warning state, wherein in the embodiment, the relationship between the system variables, the SOC value and the power station SOC value in the four states of the charging warning state, the charging early warning state, the discharging warning state and the discharging early warning state is shown in Table 1, and the specific classification process is as follows:

[0044] Table 1 Relationship between system variables, SOC value and power station SOC value in four states

[0045]

[0046] Among them, is the current energy storage system output power, E is the total energy of the current energy storage power station system;

[0047] S3, filtering and calculating the moving average frequency of the grid frequency, and then classifying the grid frequency into a second number of levels; in the embodiment, the second number is three, including a high state, a normal state and a low state, and the specific classification process is as follows:

[0048] The frequency filtering process includes: according to the obtained grid frequency, calculating the moving average frequency according to a first preset time length In the embodiment, the first preset time length is 10 seconds, and the moving average frequency The calculation formula of the moving average frequency is:

[0049] ,

[0050] Among them, is the moving average frequency, with the unit of Hz; is the real-time frequency value of the i th sampling, with the unit of Hz; N represents the number of sampling points, which is determined according to the preset sampling frequency and the first preset time length, in the embodiment, 100 points are sampled every 100 milliseconds, so N is 100;

[0051] Based on the calculated frequency moving average value Classification is carried out;

[0052] (1) When the frequency interval is [50.1, 50.5], the frequency is in the "high" state;

[0053] (2) When the frequency interval is [49.9, 50.1], the frequency is in the "normal" state.

[0054] (3) When the frequency interval is [49.9, 49.5], the battery is in the "low" state;

[0055] S4, based on the frequency moving average Calculate the frequency difference value , the calculation formula is as follows:

[0056] ,

[0057] Wherein, is a preset frequency reference value, when the frequency interval of the real-time acquired power grid frequency is [49.9, 50.1], the preset frequency reference value is 50; when the frequency interval is [50.1, 50.5], the preset frequency reference value is 50.1; when the frequency interval is [49.9, 49.5], the preset frequency reference value is 49.9;

[0058] S5, based on the classification of SOC and frequency, combined with the frequency difference value to realize frequency control; the process includes the following contents:

[0059] (1) The battery is in normal state, the energy storage system outputs linearly according to the frequency change, and the linear parameters are set according to the frequency classification , [49.9, 50.1] is set as the first frequency region, [49.5, 49.9] and [50.1, 50.5] are set as the second frequency region, the system is in the first frequency region and the first frequency region, and the first current energy storage system should output power according to the first current energy storage system should output power, and the system outputs according to the maximum frequency when the frequency moving average is lower than 49.5 and higher than 50.5. The first current energy storage system should output power is calculated by linear parameters , so that the control granularity of the first frequency region is smaller, and the control granularity of the second frequency region is larger. The first current energy storage system should output power is:

[0060] ,

[0061] Wherein, is the current energy storage system should be power; when the frequency interval is [49.9, 50.1], that is, the frequency is in the "normal" state, , when the frequency is in other states, , is preset frequency reference value, different fixed value is set for different frequency interval; different linear parameter is set according to different frequency interval , the stability of power grid frequency can be better guaranteed.

[0062] (2) the battery is in charging prewarning, discharging prewarning state; the energy storage system linearly outputs according to frequency variation, and simultaneously allows to compensate battery power, prewarning state compensation coefficient is set according to frequency grading, and the second current energy storage system output power is calculated in combination with the first current energy storage system output power , to ensure longer frequency modulation service. To ensure smooth compensation, compensation power should be smoothly provided according to the time of SOC entering prewarning state in the past 5 minutes, so that the prewarning state variable is calculated once every second, and there are 300 points, and the prewarning state rolling average value at the current time is calculated, and the calculation formula is as follows:

[0063] ,

[0064] wherein, is the SOC value of the discharging prewarning state in the SOC descending process, i.e. the variable SOC enable warning, lower in the SOC descending process and SOC disalbewarning, lower in the SOC ascending process in table 1, is the SOC value of the charging prewarning state in the SOC ascending process, i.e. the variable SOC enable warning, upper in the SOC ascending process and SOC disalbewarning, upper in the SOC descending process in table 1, is the current time to the nth point in the 300 time points in the past 5 minutes, when being in the charging prewarning state, , indicates that the system is in high prewarning state, when the battery is in the discharging prewarning state, , indicates that the system is in low prewarning state, otherwise 0, indicating that the system is not in prewarning state.

[0065] The calculation formula of the second current energy storage system output power is as follows:

[0066] ,

[0067] wherein, The pre-alarm state compensation coefficient is set according to frequency grading, different frequency intervals are set with different values, the normal state of frequency is set as 0.3, and other states are set as 0.2.

[0068] In the SOC pre-alarm state, the appropriate compensation mechanism can provide the utilization rate of the energy storage system while providing the frequency modulation capability, and the sampling moving average algorithm can reduce the impact on the power grid.

[0069] (3) When the battery is in the charging alarm and discharging alarm state, that is, or The energy storage system only provides instantaneous adjustment when the frequency changes, and exits after the frequency stabilizes. In order to ensure stable exit, the 5-minute moving average of the frequency change is calculated first , and the calculation formula is:

[0070] ,

[0071] Among them, is the frequency moving average value of the nth point in the 300 time points from the current time to the previous 5 minutes.

[0072] The third current energy storage system output power The calculation formula is:

[0073] ,

[0074] In the SOC alarm state, the frequency modulation capability provided by the energy storage is very limited, so a dynamic response algorithm is adopted to provide frequency modulation power when the frequency changes, and to exit smoothly. In this way, even in the case of SOC alarm, frequency modulation capability can be provided, the utilization rate of energy storage can be maximized, and the impact on the power grid can be reduced.

[0075] In other embodiments of the present application, a computer device is also disclosed, which comprises a processor and a memory, the memory stores a computer program, and the processor is used to execute the computer program to implement the above-mentioned energy storage-based fast and reliable frequency modulation method. The memory can be various types of memory, which can be random access memory, read-only memory, flash memory, etc.

[0076] In addition, another embodiment of the present application also discloses a computer storage medium which stores a computer program, and the computer program is executed on a processor to implement the above-mentioned energy storage-based fast and reliable frequency modulation method. The computer storage medium can be a readable storage medium, a non-volatile storage medium or a volatile storage medium. For example, the computer storage medium can include but is not limited to various program code storage media such as a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk. The method disclosed in the present application can also be implemented by a system hardware, and the system hardware platform is equipped with an ARM architecture 4-core Cortex-A55 processor (main frequency 1.8 GHz), 4 GB memory (RAM) and 32 GB storage (ROM), 11 RS485, 4 network interfaces and 2 CAN. The operating system is linux.

[0077] The above-mentioned is only the preferred embodiment of the present application, and it should be pointed out that the ordinary skilled in the art can make several improvements and refinements without departing from the principle of the present application, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. A fast and reliable frequency regulation method based on energy storage, characterized in that, The methods include: The grid frequency is obtained in real time through a frequency detection device, and the SOC value of the battery in the energy storage system is also obtained in real time. Based on the obtained SOC value, the battery status is classified into the first quantity level. The power grid frequency is filtered and the moving average frequency is calculated. Then, the power grid frequency is classified into a second order of magnitude. Calculate the frequency difference based on the frequency moving average; Frequency modulation control is achieved by combining SOC classification and frequency classification with frequency difference. Frequency modulation control is achieved based on SOC and frequency classification, combined with frequency difference, including: When the battery is in normal condition, the energy storage system outputs power linearly according to frequency changes, and linear parameters are set according to frequency levels. [49.9, 50.1] is set as the first-level frequency region, and [49.5, 49.9] and [50.1, 50.5] are set as the second-level frequency region. When the system is in the first-level frequency region, it outputs power according to the first current energy storage system. When the frequency moving average is lower than 49.5 and higher than 50.5, the system outputs at the maximum frequency. When the battery is in a charging warning or discharging warning state, the energy storage system outputs power linearly according to the frequency change, while allowing compensation for the battery capacity. The warning state compensation coefficient is set according to the frequency level, and the second current energy storage system output power is calculated by combining the first current energy storage system output power. When the battery is in a charging alarm or discharging alarm state, the energy storage system only provides instantaneous adjustment when the frequency changes, and stops after the frequency stabilizes; The formula for calculating the output power of the first current energy storage system includes: ， in, This represents the current required power output of the energy storage system. These are linear parameters, set according to frequency levels. This is the frequency difference; The formula for calculating the output power of the second current energy storage system includes: ， in, Current time Within the first 5 minutes, the warning status variable at the nth point out of 300 time points, when in a charging warning state, This indicates that the system is in a high-alert state. This occurs when the battery is in a discharge alert state. A value of 0 indicates that the system is in a low-alert state; otherwise, a value of 0 indicates that the system is not in an alert state. The compensation coefficient for the early warning status is set according to the frequency classification. This represents the current required power output of the energy storage system. The first current energy storage system should output power; Current time The rolling average of the warning status; The formula for calculating the output power of the third current energy storage system includes: ， in, Current time The frequency moving average of the nth point out of 300 time points within the previous 5 minutes. The moving average frequency, This represents the current required power output of the energy storage system. The frequency change is a 5-minute moving average.

2. The fast and reliable frequency regulation method based on energy storage according to claim 1, characterized in that: Frequency filtering includes: calculating a moving average frequency based on the obtained power grid frequency and a first preset length. Moving average frequency The calculation formulas include: ， in, It is the moving average frequency; is the real-time frequency value of the i-th sample; N represents the number of sampling points.

3. The fast and reliable frequency regulation method based on energy storage according to claim 1, characterized in that: The first number is five, including charging alarm status, charging warning status, normal status, discharge alarm status, and discharge warning status.

4. The fast and reliable frequency regulation method based on energy storage according to claim 1, characterized in that: The second set of quantities consists of three states: high, normal, and low; based on the calculated frequency moving average. The frequency is classified as follows: when the frequency range is [50.1, 50.5], the frequency is in a "high" state; when the frequency range is [49.9, 50.1], the frequency is in a "normal" state; when the frequency range is [49.9, 49.5], the battery is in a "low" state.

5. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement a fast and reliable frequency modulation method based on energy storage as described in any one of claims 1 to 4.

6. A computer storage medium, characterized in that, It stores a computer program, which, when executed on a processor, implements a fast and reliable frequency modulation method based on energy storage according to any one of claims 1 to 4.

Images