Power system power demand compensation control method under energy storage output power constraint

Abstract

The invention relates to the technical field of electric power control, and particularly discloses an electric power system power demand compensation control method under energy storage output power constraint, and the method comprises the steps: constructing a load frequency control model of an interconnected electric power system between a first region and a second region, and carrying out the power transmission and exchange between the regions through a tie line, and establishing a state-space equation containing the output power limitation of the energy storage system, and converting the frequency deviation and the tie line power deviation into error signals by using a preset performance function so as to control the charging and discharging power of the energy storage system. Through the mode, the stable frequency of the power system can be ensured, the output power of the energy storage system can be effectively limited, and the situation that the energy storage system outputs over-design power due to sudden change of the frequency is avoided, so that internal components of the energy storage system are protected, and the service life of the energy storage system is prolonged.

Description

Technical Field

[0001] The present application relates to the field of power control technology, and more specifically, to a power system power demand compensation control method under energy storage output power constraints. Background Art

[0002] With the rapid development of the global economy, the scale of the power system, as a key infrastructure of modern society, is expanding at an unprecedented rate. The load demand it carries is also rising, and its frequency stability problem is becoming more and more critical. The frequency stability of the power system depends on the dynamic balance between load power and power generation power. Once the imbalance causes the frequency deviation of the power grid, it will seriously threaten the safe operation of the system. Power failures have occurred in many places around the world due to frequency deviation.

[0003] In order to solve the uneven distribution of load and generation resources, multi-regional power systems use interconnection lines to allocate resources. This method alleviates the contradiction of uneven distribution of generation resources to a certain extent, but the load changes in each region will still have an adverse impact on the global frequency, and the power deviation of the interconnection line will also significantly affect the overall stability of the power system. Load frequency control (LFC), as one of the core functions of automatic generation control, plays a key role in maintaining the balance between generation power and load power and maintaining system frequency stability. Existing LFC control strategies include PID controller based on gray wolf optimization algorithm, control strategy of MVO algorithm and improved objective function, control method combining IPOA and PIDD2-FOI, and fractional-order PID controller. Although these methods can converge the frequency deviation to zero to a certain extent, there are still problems such as excessive transient response overshoot and long convergence time, and the controller output amplitude fluctuates greatly, which is not easy to meet the needs of safe and efficient operation of the power system. In addition, when the energy storage system participates in frequency control, if the input signal has no saturation limit, the sudden change of the power system frequency will cause the energy storage system to output over-design power, resulting in damage to internal components and accelerated aging, thereby affecting the stability of the power system.

[0004] Therefore, an optimized power system power demand compensation control method under the energy storage output power constraint is expected. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. An embodiment of the present application provides a method for compensating and controlling power demand of an electric power system under the constraint of energy storage output power, which constructs a load frequency control model of an interconnected electric power system between a first area and a second area, uses interconnection lines to perform power transmission and exchange between areas, and establishes a state space equation containing the output power limit of the energy storage system, and uses a preset performance function to convert the frequency deviation and the interconnection line power deviation into an error signal, thereby controlling the charging and discharging power of the energy storage system. In this way, while ensuring the stability of the frequency of the electric power system, the output power of the energy storage system can be effectively limited to prevent it from outputting over-design power due to sudden changes in frequency, thereby protecting the internal components of the energy storage system and extending its service life.

[0006] Accordingly, according to one aspect of the present application, a method for compensating and controlling power demand of an electric power system under energy storage output power constraints is provided, which comprises: Step 1: constructing a load frequency control model of the interconnected power system between the first region and the second region to achieve coordinated frequency deviation regulation between the first region and the second region through the tie line; Step 2: constructing a state space equation of the first region, wherein the state space equation is constructed based on a system state quantity matrix, an input matrix and a disturbance matrix, and each energy storage input signal in the input matrix is ​​less than the energy storage output power; Step 3: extracting frequency deviation and tie line power deviation from the system state quantity matrix, and determining a control input signal based on the frequency deviation and tie line power deviation, wherein the control input signal is used to control the charging and discharging power of the energy storage system.

[0007] Compared with the prior art, the power demand compensation control method of the power system under the energy storage output power constraint provided by the present application constructs a load frequency control model of the interconnected power system between the first area and the second area, uses the interconnection line to perform power transmission and exchange between the areas, and establishes a state space equation containing the output power limit of the energy storage system, and uses a preset performance function to convert the frequency deviation and the interconnection line power deviation into an error signal, thereby controlling the charging and discharging power of the energy storage system. In this way, while ensuring the stability of the power system frequency, the output power of the energy storage system can be effectively limited to avoid outputting over-design power due to sudden frequency changes, thereby protecting the internal components of the energy storage system and extending its service life. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0009] Figure 1 Flow chart of a method for controlling power demand compensation of a power system under energy storage output power constraints according to an embodiment of the present application.

[0010] Figure 2 A load frequency control model diagram of an interconnected power system according to an embodiment of the present application.

[0011] Figure 3 Graph showing the physical meaning of parameters of the state space equation of the first region according to an embodiment of the present application.

[0012] Figure 4 This is a flowchart of step S3 in the power system power demand compensation control method under the energy storage output power constraint according to Example 2 of the present application.

[0013] Figure 5 This is a flowchart of step S33B in the power system power demand compensation control method under the energy storage output power constraint according to Example 2 of the present application. DETAILED DESCRIPTION

[0014] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0015] Example 1 As mentioned in the background technology above, the frequency deviation of the power grid has an important impact on the safe and stable operation of the power system. For motors, frequency deviation will cause their speed to be unstable, affecting production efficiency, and may also cause the motor to overheat and be damaged; for transformers, frequency deviation will change their excitation current, increase iron loss, reduce operating efficiency, and may also shorten their service life in the long run; severe frequency deviation may even cause power system failures, trigger emergency shutdowns of multiple generators, and affect the continuous supply of electricity.

[0016] Therefore, in order to achieve effective compensation for the power demand of the power system, the present invention proposes a power demand compensation control method for the power system under the constraint of energy storage output power, which constructs a load frequency control model of the interconnected power system between the first area and the second area, uses the interconnection line to perform power transmission and exchange between regions, and establishes a state space equation containing the output power limit of the energy storage system, and uses a preset performance function to convert the frequency deviation and the interconnection line power deviation into an error signal, so as to control the charging and discharging power of the energy storage system. In this way, while ensuring the stability of the power system frequency, the output power of the energy storage system can be effectively limited to avoid outputting over-design power due to sudden frequency changes, thereby protecting the internal components of the energy storage system and extending its service life.

[0017] Figure 1 Flow chart of the power demand compensation control method of the power system under the energy storage output power constraint according to the embodiment of the present application. Figure 1 As shown, according to the power demand compensation control method of the power system under the energy storage output power constraint of the embodiment of the present application, the method includes the following steps: Step 1: constructing a load frequency control model of the interconnected power system between the first area and the second area to achieve coordinated regulation of the frequency deviation between the first area and the second area through the interconnection line; Step 2: constructing the state space equation of the first area, the state space equation is constructed based on the system state quantity matrix, the input matrix and the disturbance matrix, and each energy storage input signal in the input matrix is ​​less than the energy storage output power; Step 3: extracting the frequency deviation and the interconnection line power deviation from the system state quantity matrix, and determining the control input signal based on the frequency deviation and the interconnection line power deviation, and the control input signal is used to control the charging and discharging power of the energy storage system.

[0018] In the above-mentioned power system power demand compensation control method under the energy storage output power constraint, the step 1: construct a load frequency control model of the interconnected power system between the first area and the second area to achieve coordinated adjustment of the frequency deviation between the first area and the second area through the interconnection line. It should be understood that the scale of the power system is becoming increasingly large, and there are differences in the power load characteristics and power generation resource distribution in different regions. When facing load fluctuations, a single area has limited self-regulation capabilities and it is difficult to maintain a stable frequency. By interconnecting regions and building a load frequency control model, the power generation and load resources of different regions can be integrated to improve the reliability and stability of the system. In this way, the output signal of each area is transmitted to another area through the interconnection line, so as to coordinately adjust the frequency deviation between the two areas. When a large load change occurs in a certain area, the other area provides power support in time through the interconnection line to avoid excessive frequency fluctuations in the area.

[0019] Further, Figure 2FIG. 1 is a load frequency control model diagram of an interconnected power system according to an embodiment of the present application. Figure 2 As shown, each control area includes key control units, such as controllers, energy storage systems, speed governors, reheat steam turbine units and generator sets; the key control units play their own unique roles in the dynamic regulation of system frequency deviations, and achieve system frequency stability and load balance through multi-level coordinated control.

[0020] In the above-mentioned power system power demand compensation control method under the energy storage output power constraint, the step 2: construct the state space equation of the first area, the state space equation is constructed based on the system state quantity matrix, the input matrix and the disturbance matrix, and each energy storage input signal in the input matrix is ​​less than the energy storage output power. That is, using the state space equation to describe the relationship between the state change of the system and the input and output helps to deeply understand the dynamic characteristics of the power system and provide a theoretical basis for the subsequent controller design. Specifically, the state space equation regards the power system as a dynamic state set, and describes the current state of the system through the system state quantity matrix, such as the speed of the generator, the frequency deviation, the tie line power deviation, etc. The input matrix contains the input signal of the system, which is mainly the signal for controlling the charging and discharging of the energy storage system in this application. The disturbance matrix is ​​used to consider various uncertain factors in the power system, such as random changes in loads, failures of power generation equipment, etc. At the same time, by limiting the various energy storage input signals in the input matrix to be less than the energy storage output power, the safe and stable operation of the energy storage system can be ensured, and damage or performance degradation of the energy storage equipment caused by excessive charging and discharging can be avoided.

[0021] Specifically, the state space equations and state for: ; in, and are the state quantity, output and input of the system respectively; , , and They are the state matrix, input matrix, disturbance matrix and output matrix of the system respectively. The specific forms are as follows: ; Among them, the physical meaning of each parameter in the state space equation of the first region is as follows Figure 3 shown.

[0022] In the above-mentioned power system power demand compensation control method under the energy storage output power constraint, the step 3: extract the frequency deviation and the interconnection line power deviation from the system state quantity matrix, and determine the control input signal based on the frequency deviation and the interconnection line power deviation, and the control input signal is used to control the charging and discharging power of the energy storage system. It should be understood that the frequency deviation directly reflects the degree of imbalance between the power generation power and the load power, and the interconnection line power deviation reflects the abnormal situation of the power exchange between regions. Both are important bases for guiding the charging and discharging operation of the energy storage system. Therefore, the present application further combines the frequency deviation and the interconnection line power deviation to calculate the error signal, and determines the control input signal accordingly, and adjusts the charging and discharging power of the energy storage system, so as to balance the power generation power and the load power, reduce frequency fluctuations, and maintain the stability of the power exchange between regions.

[0023] Specifically, the step 3 includes: step 31A: calculating an error signal based on the frequency deviation and the tie line power deviation; step 32A: determining the control input signal based on the error signal.

[0024] More specifically, in step 31A, first, the frequency deviation and the tie line power deviation are accumulated to obtain an original error signal. Here, the frequency deviation focuses on the balance of power within the system, and the tie line power deviation focuses on power coordination between regions. By accumulating the frequency deviation and the tie line power deviation as the original error signal, the system's needs in terms of power allocation and frequency regulation can be comprehensively considered, and the current overall imbalance of the system can be more comprehensively reflected, thereby providing a comprehensive and representative error benchmark for subsequent control adjustments.

[0025] Next, the boundedness analysis of the original error signal is performed using the boundary function to obtain a verified conversion error signal. That is, considering that in the actual power system control, the value range of the error signal needs to be reasonably limited to ensure the stability and safety of the system. If the error signal is too large, it may cause the control signal to exceed the range that the system can bear, thereby causing problems such as system out of control or equipment damage. Based on this, the present application further uses the boundary function to perform a boundedness analysis on the original error signal. When the original error signal exceeds the boundary function value, it is adjusted according to certain rules to obtain a verified conversion error signal. In a specific example of the present application, the boundary function is an exponential decay boundary function. Specifically, the exponential decay boundary function gradually weakens the limiting effect on the error signal over time. In the initial stage of the system, the exponential decay boundary function can quickly limit the excessive error signal to a safe range. As the system gradually stabilizes, it will not over-limit the change of the error signal, thereby ensuring that the system has a certain dynamic response capability. In this way, the potential harm caused by the excessive error signal to the system is effectively avoided, and the system is guaranteed to operate within a safe range.

[0026] Then, the verification conversion error signal is converted using the error conversion equation to obtain the error signal. That is, in order to achieve smooth conversion of the error signal and avoid drastic fluctuations in the control signal due to sudden changes in the error signal, the present application further uses the error conversion equation to convert the verification conversion error signal. In a specific example of the present application, the error conversion equation is a hyperbolic tangent function or a logarithmic function. It should be understood that both the hyperbolic tangent function and the logarithmic function have the ability to map a wide range of inputs to a limited output range, and the output changes are relatively smooth, which can effectively suppress the impact of sudden changes in the error signal on the control signal. Through the processing of the error conversion equation, the error signal obtained not only retains the main characteristics of the original error signal, but also has better smoothness and controllability, providing a more stable and reliable basis for the subsequent control strategy design.

[0027] Finally, in step 32A, the error signal is input into a controller to obtain the control input signal, and the controller uses a Lyapunov function, an error dynamics equation, and a feedback linearization control model to determine the control input signal.

[0028] First, based on the preset performance control idea, the error The transient process control is transformed into a finite-dimensional input design that meets the preset performance indicators; then, the control input signal is determined , and Substituted as input into the auxiliary system to control the charging and discharging power of the energy storage system; Furthermore, in the signal processing process, the saturation function The limiting function is to intercept the control signal that exceeds the range to the upper or lower limit allowed, thereby preventing the input signal from exceeding the capacity limit of the energy storage system; Furthermore, in the above-mentioned power system power demand compensation control method under the energy storage output power constraint, it also includes step 4: obtaining the control signal Input the system and continuously detect the error signal; if the error signal is within the allowable fluctuation range, the process ends; if the error signal is not within the allowable fluctuation range, steps 1 to 4 are executed repeatedly until the system output is stable; Furthermore, considering the characteristics of the actual system, the input amplitude is constrained by the saturation function, that is: ; Simplify the calculation process, let .

[0029] Furthermore, the design of the preset performance controller under the premise of input saturation constraint includes five processes: defining the error function, selecting the boundary function, error conversion, anti-saturation design, and controller derivation; the specific steps are as follows: Define the error function: Area control error The goal of ACE control is to make , and set the reference value to ; At time t, the expected system output is Can be compared with the reference value Keep consistent; the error function can be expressed as: ; Select boundary function: Based on the PPC idea, the error The boundedness analysis of is converted into the verification conversion error Asymptotic stability problems in Lyapunov's law; choose As the boundary function of PPC; Error conversion: The error conversion equation is used to convert the error. ; right The time is derived and designed to obtain ; in, ; Finally, the error of the definition error is ; Anti-saturation design: middle Designed as follows ; According to the Thus there is The time derivative of ; Controller derivation: Define the Lyapunov function as ; Taking its derivative, we get ; Derive the controller based on the backstepping method ; Substituting it into the Lyapunov function, we get ; Bounded, on the other hand, the system It belongs to a bounded input and bounded state (BIBO) system, so , is also bounded; according to the PPC conclusion Under the premise of being bounded, Can be constrained within a preset range.

[0030] In summary, according to the embodiment of the present application, the power demand compensation control method of the power system under the energy storage output power constraint is explained, which constructs a load frequency control model of the interconnected power system between the first area and the second area, uses the interconnection line to perform power transmission and exchange between regions, and establishes a state space equation containing the output power limit of the energy storage system, and uses a preset performance function to convert the frequency deviation and the interconnection line power deviation into an error signal, so as to control the charging and discharging power of the energy storage system. In this way, while ensuring the stability of the power system frequency, the output power of the energy storage system can be effectively limited to avoid outputting over-design power due to sudden frequency changes, thereby protecting the internal components of the energy storage system and extending its service life.

[0031] Example 2 In particular, considering that the power system is a complex system that operates dynamically, its frequency deviation and interconnection line power deviation fluctuate continuously over time. If only the deviation value at a certain moment is focused on to control the charging and discharging power of the energy storage system, it may be affected by the dynamic characteristics of the system and cannot effectively cope with the rapid changes of the system, resulting in unsatisfactory control effects. For example, when the power system suffers a sudden load shock, the frequency deviation and interconnection line power deviation will show complex fluctuations over a period of time. Control based only on the deviation value at a single moment may cause delayed or inaccurate control decisions. Therefore, the present application proposes a charging and discharging power control method based on deep learning technology, which performs time-series interactive analysis on the frequency deviation and the interconnection line power deviation to capture the time-series fluctuation characteristics and dynamic correlation patterns of the two, so as to deeply understand the time-series trend of the deviation and make reasonable charging and discharging power control decisions accordingly.

[0032] Figure 4 FIG. 1 is a flow chart of step S3 in the power system power demand compensation control method under the energy storage output power constraint according to Example 2 of the present application. Figure 4 As shown, the step S3 includes: step 31B: extracting the time queue of the frequency deviation and the time queue of the tie line power deviation from the system state quantity matrix; step 32B: extracting the deviation timing fluctuation characteristics from the time queue of the frequency deviation and the time queue of the tie line power deviation to obtain the frequency deviation timing fluctuation characteristic coding vector and the tie line power deviation timing fluctuation characteristic coding vector; step 33B: performing feature interaction based on sparse constraints on the frequency deviation timing fluctuation characteristic coding vector and the tie line power deviation timing fluctuation characteristic coding vector to obtain the frequency deviation-tie line power deviation timing interaction aggregation characteristic coding vector; step 34B: determining the control input signal based on the frequency deviation-tie line power deviation timing interaction aggregation characteristic coding vector.

[0033] Specifically, the step 31B: extracting the time queue of the frequency deviation and the time queue of the tie line power deviation from the system state quantity matrix. Here, by extracting the time queue of the frequency deviation and the time queue of the tie line power deviation, complete information of the frequency deviation and the tie line power deviation changing with time can be obtained, thereby providing basic data for the control decision of the charging and discharging power.

[0034] Specifically, step 32B: extracting deviation time series fluctuation characteristics from the time queue of the frequency deviation and the time queue of the tie line power deviation to obtain frequency deviation time series fluctuation characteristic encoding vector and tie line power deviation time series fluctuation characteristic encoding vector. That is, in order to capture the fluctuation trend of frequency deviation and tie line power deviation in the time dimension, the present application further adopts a deep learning model to perform time series modeling analysis on the time queue of the frequency deviation and the time queue of the tie line power deviation. In a specific example of the present application, a long short-term memory network (LSTM) is used to extract characteristics of the time queue of the frequency deviation and the time queue of the tie line power deviation. As a special recurrent neural network (RNN), the long short-term memory network (LSTM) has excellent long-term time series dependent information memory ability by introducing structures such as forget gates, input gates and output gates. In the present application, through the processing of the LSTM network, the amplitude, periodicity and change rate of the deviation time series fluctuation can be effectively extracted from the time series of the frequency deviation and the tie line power deviation, and encoded into a frequency deviation time series fluctuation characteristic encoding vector and a tie line power deviation time series fluctuation characteristic encoding vector.

[0035] Specifically, step 33B: performs feature interaction based on sparse constraints on the frequency deviation time series fluctuation feature coding vector and the interconnection line power deviation time series fluctuation feature coding vector to obtain a frequency deviation-interconnection line power deviation time series interaction aggregation feature coding vector. That is, considering that in the power system, the frequency deviation and the interconnection line power deviation are not independent variables, and there is a complex interactive relationship between the two. For example, when the load in a certain area changes, it will not only cause the frequency deviation of the area, but also affect the power distribution of other areas through the interconnection line, thereby causing a change in the interconnection line power deviation. At the same time, the adjustment of the interconnection line power will also have a feedback effect on the frequency of the system. However, the traditional method lacks the analysis of the timing interaction between the two, which may lead to the control strategy being unable to effectively coordinate the influence between the two, reducing the overall control effect of the system. In this regard, the present application proposes a feature interaction method based on sparse constraints, which reconstructs the feature phase space of the frequency deviation timing fluctuation feature coding vector and the interconnection line power deviation timing fluctuation feature coding vector, so as to utilize the implicit query space to characterize the complex interaction pattern between features, and imposes spatial sparse constraints to reduce the interference of redundant information and highlight key interaction features, thereby obtaining the frequency deviation-interconnection line power deviation timing interaction aggregated feature coding vector, providing richer information basis for subsequent charging and discharging power control decisions.

[0036] Figure 5 FIG. 2 is a flow chart of step S33B in the power system power demand compensation control method under the energy storage output power constraint according to Example 2 of the present application. Figure 5 As shown, the step 33B includes: step 331: reconstructing the characteristic phase space of the frequency deviation time series fluctuation characteristic coding vector and the tie line power deviation time series fluctuation characteristic coding vector based on one-dimensional convolution coding to obtain a set of frequency deviation local time series fluctuation characteristic coding vectors and a set of tie line power deviation local time series fluctuation characteristic coding vectors; step 332: based on the implicit query space correlation strength between each group of corresponding frequency deviation local time series fluctuation characteristic coding vectors and tie line power deviation local time series fluctuation characteristic coding vectors in the set of frequency deviation local time series fluctuation characteristic coding vectors and the set of tie line power deviation local time series fluctuation characteristic coding vectors, sparse constrained aggregation is performed on the set of frequency deviation local time series fluctuation characteristic coding vectors and the set of tie line power deviation local time series fluctuation characteristic coding vectors to obtain the frequency deviation-tie line power deviation time series interaction aggregation characteristic coding vector.

[0037] In a specific example of the present application, step 331 is expressed by the formula: in, represents the frequency deviation timing fluctuation characteristic encoding vector, represents the characteristic coding vector of the time series fluctuation of the tie line power deviation, is a one-dimensional convolutional coding network, Represents the set of encoding vectors of the local timing fluctuation characteristics of the frequency deviation, , , and respectively represent the first, second, and third frequency deviation local time series fluctuation feature encoding vectors in the set and A frequency deviation local time series fluctuation feature encoding vector, is the number of vectors in the set of the frequency deviation local timing fluctuation feature encoding vectors, represents the set of encoding vectors of local time series fluctuation characteristics of the tie line power deviation, , , and respectively represent the first, second, and third in the set of the local time series fluctuation characteristic coding vectors of the tie line power deviation and The local time series fluctuation characteristic encoding vector of the interconnection line power deviation is constructed.

[0038] That is, the frequency deviation time series fluctuation feature coding vector and the interconnection line power deviation time series fluctuation feature coding vector are regarded as some potential tracks in the abstract feature space, and the two are locally perceived in the form of sliding windows through one-dimensional convolution coding to capture the local feature patterns at different time series positions and enhance the perception of local structural information. Thus, through the parallel operation of multiple groups of convolution kernels, a multi-perspective representation of the original features in different modes is formed, and a set of frequency deviation local time series fluctuation feature coding vectors and a set of interconnection line power deviation local time series fluctuation feature coding vectors are obtained. It is worth mentioning here that the set of frequency deviation local time series fluctuation feature coding vectors and the set of interconnection line power deviation local time series fluctuation feature coding vectors also constitute a kind of feature redundancy. This redundancy is not invalid information, but a repeated representation of the original features at different local perspectives, which helps to improve the model's anti-noise ability.

[0039] In a specific example of the present application, the step 332 includes: first, calculating the implicit query space joint coding matrix between each corresponding frequency deviation local time series fluctuation feature coding vector and the tie line power deviation local time series fluctuation feature coding vector in the set of the frequency deviation local time series fluctuation feature coding vector and the tie line power deviation local time series fluctuation feature coding vector to obtain a set of frequency deviation-tie line power deviation timing interaction implicit query space joint coding matrices, which is expressed by the formula: in, represents the transpose of a matrix, represents the matrix multiplication operation, and They represent the characteristic weight matrix of frequency deviation time series fluctuation and the characteristic weight matrix of tie line power deviation time series fluctuation, respectively. is the characteristic scale scaling factor, express and The frequency deviation-tie line power deviation timing interaction implicitly queries the space joint coding matrix.

[0040] That is, by performing implicit query space mapping and feature interaction coding on each group of corresponding frequency deviation local time series fluctuation feature coding vectors and tie line power deviation local time series fluctuation feature coding vectors, the potential correlation patterns of the two at different time series positions are mined, and the pairwise interaction relationships between their different dimensions are constructed in the implicit query space to obtain a set of frequency deviation-tie line power deviation time series interaction implicit query space joint coding matrices. Here, the implicit query space joint coding matrix serves as the carrier of the time series correlation interaction information between the frequency deviation and the tie line power deviation, providing an information basis for subsequent sparse constraints and adaptive aggregation.

[0041] Then, the spatial sparse constraint factor of each frequency deviation-tie line power deviation timing interaction implicit query space joint coding matrix in the set of the frequency deviation-tie line power deviation timing interaction implicit query space joint coding matrix is ​​calculated to obtain a set of frequency deviation-tie line power deviation timing interaction space sparse constraint factors, which is expressed by the formula: in, is the correlation feature strength measurement function, It means to calculate the square of the Frobenius norm of the matrix. express The corresponding frequency deviation-interconnection line power deviation timing interaction spatial sparse constraint factor.

[0042] Due to the above-mentioned feature phase space reconstruction process, the set of the frequency deviation-interconnection line power deviation timing interaction implicit query space joint coding matrix contains a large amount of redundant information and noise. Therefore, in order to reduce the amount of calculation and reduce information interference, the present application further identifies the degree of feature importance and calculates its spatial sparse constraint factor by performing feature distribution statistical analysis on each frequency deviation-interconnection line power deviation timing interaction implicit query space joint coding matrix to obtain a set of frequency deviation-interconnection line power deviation timing interaction space sparse constraint factors. Here, the frequency deviation-interconnection line power deviation timing interaction space sparse constraint factor is the square value of the Frobenius norm of the frequency deviation-interconnection line power deviation timing interaction implicit query space joint coding matrix.

[0043] Furthermore, based on the set of sparse constraint factors of the frequency deviation-tie line power deviation time series interaction space, the set of the frequency deviation-tie line power deviation time series interaction implicit query space joint coding matrices is adaptively aggregated and coded to obtain the frequency deviation-tie line power deviation time series interaction aggregate feature coding vector, which is expressed as follows: in, represents the normalized exponential function, express The corresponding normalized frequency deviation-interconnection line power deviation timing interaction spatial sparse constraint factor, Represents the frequency deviation-interconnection line power deviation time series interaction implicit coding feature sparsity fusion matrix, represents the feature shape reshaping, Represents the frequency deviation-interconnection line power deviation time series interactive aggregation feature coding vector.

[0044] Here, the set of sparse constraint factors of the frequency deviation-tie line power deviation timing interaction space reflects the importance of each frequency deviation-tie line power deviation timing interaction implicit query space joint coding matrix in the overall feature interaction, and provides a basis for subsequent adaptive aggregation. Through normalization processing, the set of sparse constraint factors of the frequency deviation-tie line power deviation timing interaction space is converted into a weight distribution, so as to perform weighted aggregation and feature shape reshaping on the set of frequency deviation-tie line power deviation timing interaction implicit query space joint coding matrices, and obtain the frequency deviation-tie line power deviation timing interaction aggregated feature coding vector after sparse constraint, which effectively realizes the suppression of redundant information, the strengthening of key interaction features, and improves the discrimination and effectiveness of feature representation.

[0045] Specifically, the step 34B: based on the frequency deviation-interconnection line power deviation timing interaction aggregation feature coding vector, determine the control input signal. In an embodiment of the present application, a decoder structure based on a neural network model is used to decode the frequency deviation-interconnection line power deviation timing interaction aggregation feature coding vector, so as to utilize the powerful nonlinear fitting ability of the neural network to perform deep learning and modeling on the timing feature interaction pattern between the frequency deviation and the interconnection line power deviation, and predict the charging and discharging power required by the power system based on this, so as to determine the control input signal, realize adaptive control of the charging and discharging power of the energy storage device in the power system, and thus improve the stability and reliability of the power system.

[0046] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A power system power demand compensation control method under energy storage output power constraints, characterized in that: include: Step 1: constructing a load frequency control model of the interconnected power system between the first region and the second region to achieve coordinated frequency deviation regulation between the first region and the second region through the tie line; Step 2: constructing a state space equation of the first region, wherein the state space equation is constructed based on a system state quantity matrix, an input matrix and a disturbance matrix, and each energy storage input signal in the input matrix is ​​less than the energy storage output power; Step 3: extracting frequency deviation and tie line power deviation from the system state quantity matrix, and determining a control input signal based on the frequency deviation and tie line power deviation, wherein the control input signal is used to control the charging and discharging power of the energy storage system.

2. The power system power demand compensation control method under the energy storage output power constraint according to claim 1 is characterized in that: The step 3 comprises: Step 31A: Calculating an error signal based on the frequency deviation and the tie line power deviation; Step 32A: Determine the control input signal based on the error signal.

3. The power demand compensation control method for the power system under the energy storage output power constraint according to claim 2 is characterized in that: The step 31A comprises: Accumulating the frequency deviation and the tie line power deviation to obtain an original error signal; Performing boundedness analysis on the original error signal using a boundary function to obtain a verified conversion error signal; The verification conversion error signal is converted using an error conversion equation to obtain the error signal.

4. The power demand compensation control method for the power system under the energy storage output power constraint according to claim 3 is characterized in that: The boundary function is an exponential decay boundary function.

5. The power demand compensation control method for the power system under the energy storage output power constraint according to claim 3 is characterized in that: The error conversion equation is a hyperbolic tangent function or a logarithmic function.

6. The power demand compensation control method for the power system under the energy storage output power constraint according to claim 2 is characterized in that: The step 32A includes: inputting the error signal into a controller to obtain the control input signal, wherein the controller uses a Lyapunov function, an error dynamics equation and a feedback linearization control model to determine the control input signal.

7. The power demand compensation control method for the power system under the energy storage output power constraint according to claim 2 is characterized in that: Also includes step 4: continuously detecting the error signal; If the error signal is within the allowable fluctuation range, the process ends; If the error signal is not within the allowable fluctuation range, steps 1 to 4 are executed in a loop.

8. The power demand compensation control method for the power system under the energy storage output power constraint according to claim 1 is characterized in that: The step 3 comprises: Step 31B: extracting the time queue of the frequency deviation and the time queue of the tie line power deviation from the system state quantity matrix; Step 32B: extracting deviation time series fluctuation characteristics from the time queue of the frequency deviation and the time queue of the tie line power deviation to obtain a frequency deviation time series fluctuation characteristic coding vector and a tie line power deviation time series fluctuation characteristic coding vector; Step 33B: performing feature interaction based on sparse constraints on the frequency deviation time series fluctuation feature coding vector and the tie line power deviation time series fluctuation feature coding vector to obtain a frequency deviation-tie line power deviation time series interaction aggregated feature coding vector; Step 34B: Determine the control input signal based on the frequency deviation-interconnection line power deviation timing interaction aggregation feature coding vector.

9. The power demand compensation control method for the power system under the energy storage output power constraint according to claim 8 is characterized in that: The step 33B comprises: Step 331: reconstruct the characteristic phase space of the frequency deviation time series fluctuation characteristic coding vector and the tie line power deviation time series fluctuation characteristic coding vector based on one-dimensional convolution coding to obtain a set of frequency deviation local time series fluctuation characteristic coding vectors and a set of tie line power deviation local time series fluctuation characteristic coding vectors; Step 332: Based on the implicit query space correlation strength between each corresponding group of frequency deviation local timing fluctuation feature coding vectors and tie line power deviation local timing fluctuation feature coding vectors in the set of frequency deviation local timing fluctuation feature coding vectors and the set of tie line power deviation local timing fluctuation feature coding vectors, sparse constrained aggregation is performed on the set of frequency deviation local timing fluctuation feature coding vectors and the set of tie line power deviation local timing fluctuation feature coding vectors to obtain the frequency deviation-tie line power deviation timing interaction aggregation feature coding vector.

10. The power system power demand compensation control method under the energy storage output power constraint according to claim 9 is characterized in that: The step 332 includes: Calculate the implicit query space joint coding matrix between each corresponding group of frequency deviation local time series fluctuation feature coding vectors and tie line power deviation local time series fluctuation feature coding vectors in the set of the frequency deviation local time series fluctuation feature coding vectors and the set of the tie line power deviation local time series fluctuation feature coding vectors to obtain a set of frequency deviation-tie line power deviation timing interaction implicit query space joint coding matrices; Calculating the spatial sparse constraint factor of each frequency deviation-tie line power deviation timing interaction implicit query space joint coding matrix in the set of frequency deviation-tie line power deviation timing interaction implicit query space joint coding matrices to obtain a set of frequency deviation-tie line power deviation timing interaction space sparse constraint factors; Based on the set of sparse constraint factors of the frequency deviation-interconnection line power deviation timing interaction space, the set of frequency deviation-interconnection line power deviation timing interaction implicit query space joint coding matrices is adaptively aggregated and coded to obtain the frequency deviation-interconnection line power deviation timing interaction aggregate feature coding vector.

Images