New energy station voltage regulation and control optimization system based on real-time monitoring

Abstract

The invention relates to the technical field of voltage regulation and control, and discloses a new energy station voltage regulation and control optimization system based on real-time monitoring, and the system comprises a data collection module which is used for collecting power data and equipment state data in a first preset time period; the data preprocessing module is used for preprocessing the power data and the equipment state data to obtain a second feature sequence; the voltage prediction module is used for constructing a voltage prediction model according to the second feature sequence and outputting voltage in a second preset time period; and the regulation and control scheme generation module is used for generating a regulation and control scheme according to the voltage of the second preset time period by combining with a genetic algorithm which introduces adaptive t distribution disturbance variation, and determining the start and stop states of J reactive compensation devices and the positions of K voltage regulator taps in the second preset time period according to the regulation and control scheme. Through the improved voltage prediction model and the adaptive t distribution-based genetic algorithm, accurate prediction and optimal regulation and control of the power grid voltage are realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of voltage regulation, and in particular relates to a voltage regulation optimization system for a new energy station based on real-time monitoring. Background Art

[0002] As the global demand for renewable energy continues to increase, the proportion of new energy sites (such as wind farms and photovoltaic power stations) in the power system has gradually increased. However, due to the volatility and uncertainty of wind power and photovoltaic power generation, the problem of voltage fluctuation in the power grid has become increasingly prominent. Traditional voltage regulation methods often rely on static voltage regulation strategies, which cannot respond to the rapid changes in the output of renewable energy power generation in real time, resulting in frequent voltage deviations, affecting the stability and safe operation of the power grid.

[0003] The existing voltage regulation system mainly relies on reactive power compensation equipment and voltage regulators to regulate voltage, but the regulation strategies of these devices are usually based on historical data and fixed rules, lacking timely response to the dynamic characteristics of renewable energy generation, and unable to meet the current grid's demand for high-frequency voltage regulation. In addition, frequent equipment start-up and shutdown and regulation will increase system losses and maintenance costs, and reduce regulation efficiency.

[0004] Therefore, a voltage control optimization system is urgently needed to solve the above problems. Summary of the invention

[0005] The present invention provides a new energy station voltage control and optimization system based on real-time monitoring, which solves the technical problems in related technologies such as lack of real-time response capability to fluctuations in new energy power generation, untimely voltage regulation, and impact on power grid stability and operating efficiency.

[0006] The present invention provides a new energy station voltage control optimization system based on real-time monitoring, comprising: A data acquisition module is used to collect power data and equipment status data of a new energy station within a first preset time period T according to a preset time interval t, and represent it through a first characteristic sequence, wherein the nth sequence unit of the first characteristic sequence represents the power data and equipment status data at the nth time point, wherein T and t are custom parameters, N=T / t, 1≤n≤N; The power data includes: voltage, current, active power and reactive power; the equipment status data includes: the start and stop status of J reactive compensation devices and the position of K voltage regulator taps; A data preprocessing module, which is used to preprocess the first feature sequence to obtain a second feature sequence; A voltage prediction module, which is used to construct a voltage prediction model according to the second characteristic sequence, wherein the input of the voltage prediction model is the second characteristic sequence, and the output is the voltage in the second preset time period; A control scheme generating module, which is used to generate a control scheme according to the voltage in the second preset time period in combination with a genetic algorithm that introduces adaptive t-distribution disturbance variation; The start and stop states of J reactive compensation devices and the positions of K voltage regulator taps in the second preset time period are determined according to the control scheme.

[0007] Furthermore, the start and stop states of the reactive power compensation device include: enabled and disabled, and the encoding method thereof adopts binary encoding; the position of the voltage regulator tap is a discrete value and adopts integer encoding.

[0008] Furthermore, the step of preprocessing the first feature sequence includes: Step S201, using a box plot to remove outliers in the first feature sequence; Step S202, calculating the ratio of voltage to current, obtaining load change characteristics, and splicing them to the power data; Step S203, normalizing the first feature sequence by using a z-score normalization method to obtain a second feature sequence.

[0009] Furthermore, the voltage prediction model includes: a hidden layer, an extraction layer and a classifier; The hidden layer includes N first units, the nth first unit is used to input the nth sequence unit of the second feature sequence, and output a feature vector; The extraction layer is used to extract the feature vector output by the Nth first unit as an update vector; The update vector is input into a classifier, and the classification space of the classifier represents the voltage in the second preset time period.

[0010] Furthermore, the hidden layer is constructed based on an improved gated recurrent unit, and the improvements include: introducing an adjustment coefficient in an update gate, introducing a scaling factor in a reset gate, introducing a residual connection, and adding a trend gate.

[0011] Furthermore, a control scheme is generated based on the voltage prediction model combined with a genetic algorithm that introduces adaptive t-distribution disturbance variation, including the following steps: Step S301, generating an initialization population with Q number of individuals meeting the constraint conditions, where Q is a custom parameter; Step S302, constructing an objective function by combining a voltage prediction model; Step S303, obtaining the fitness value of each individual in the initialized population through objective function calculation; Step S304, selecting two individuals with the largest fitness values ​​in the initialization population as parents; Step S305, generating a random number in the range of 0 to 1, and when the random number is less than a first threshold, performing a crossover operation that satisfies the constraint condition; The crossover operation means selecting the element value of one position of the vector encoding of two individuals in the parent generation to exchange; Step S306, mutating the individuals in the population based on the adaptive t-distribution perturbation mutation strategy; Step S307, if it is determined that the current number of iterations is greater than or equal to the maximum number of iterations, the individual with the largest fitness value in the initialized population is used as the control scheme, otherwise, return to step S303, where the maximum number of iterations is a custom parameter.

[0012] Furthermore, the objective function of step S302 is: ; in represents the objective function, represents the voltage output by the voltage prediction model, Indicates the preset target voltage, Represents a custom minimum positive value, Indicates the start and stop status of the jth reactive compensation device, It indicates the adjustment amplitude of the kth voltage regulator tap relative to the previous moment.

[0013] Furthermore, the specific steps of step S306 include: Step S3061, initializing the degrees of freedom of the t distribution, wherein the probability density function of the t distribution is: ; Step S3062: dynamically adjust the degree of freedom according to the number of iterations. The degree of freedom adjustment formula is: ,in represents the current degrees of freedom, and denote the initial and final degrees of freedom, respectively. and P represent the current number of iterations and the preset maximum number of iterations respectively; Step S3063, mutate the current individual by applying t distribution to generate a new individual. The formula for generating the new individual is: ,in Represents a new individual, Represents the current individual, represents a t-distributed random number generated according to the value of the degrees of freedom, Represents the scaling factor that controls the strength of the mutation.

[0014] The beneficial effects of the present invention are as follows: the present invention constructs a voltage prediction model through an improved gated cycle unit, which can capture the complex voltage fluctuation law in the power grid, better balance the short-term fluctuation and long-term trend of the voltage, and significantly improve the accuracy and response speed of voltage prediction; The present invention uses the voltage value output by the voltage prediction model as the objective function of the optimization algorithm, so that the optimization process is directly linked to the stability of the actual voltage of the power grid. The optimization algorithm can dynamically adjust the start and stop state of the reactive compensation equipment and the tap position of the voltage regulator according to the result of the predicted voltage, making the regulation more accurate. The present invention uses a genetic algorithm that introduces an adaptive t-distribution to dynamically adjust the search range, balance the global search and the local fine search, avoid falling into the local optimum, thereby improving the global convergence ability and ensuring that the best control solution is found. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a module schematic diagram of a voltage control optimization system for a new energy station based on real-time monitoring of the present invention; Figure 2 is a flow chart of preprocessing the first feature sequence of the present invention; Figure 3 It is a flow chart of the genetic algorithm based on the adaptive t distribution strategy of the present invention; Figure 4 It is a flow chart of the individual variation based on the adaptive t distribution strategy of the present invention.

[0016] In the figure: a data acquisition module 101, a data preprocessing module 102, a voltage prediction module 103 and a control scheme generation module 104. DETAILED DESCRIPTION

[0017] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0018] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in one or more embodiments of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0019] like Figure 1-Figure 4 As shown, a voltage control and optimization system for a new energy station based on real-time monitoring includes: The data acquisition module 101 is used to collect the power data and equipment status data of the new energy station within the first preset time period T according to the preset time interval t, and represent it through the first characteristic sequence, and the nth sequence unit of the first characteristic sequence represents the power data and equipment status data at the nth time point, wherein T and t are custom parameters, N=T / t, 1≤n≤N; The power data includes: voltage, current, active power and reactive power; the equipment status data includes: the start and stop status of J reactive compensation devices and the position of K voltage regulator taps; A data preprocessing module 102, which is used to preprocess the first feature sequence to obtain a second feature sequence; A voltage prediction module 103, which is used to construct a voltage prediction model according to the second characteristic sequence, wherein the input of the voltage prediction model is the second characteristic sequence, and the output of the voltage prediction model is the voltage in the second preset time period; A control scheme generating module 104, which is used to generate a control scheme according to the voltage in the second preset time period in combination with a genetic algorithm that introduces adaptive t-distribution disturbance variation; The start and stop states of J reactive compensation devices and the positions of K voltage regulator taps in the second preset time period are determined according to the control scheme.

[0020] In one embodiment of the present invention, the start and stop states of the reactive compensation device include: enabled and disabled, and the encoding method thereof adopts binary encoding, for example, 1 represents enabled and 0 represents disabled; the position of the voltage regulator tap is a discrete value and is encoded with integers, for example, the voltage regulator tap has 17 tap positions, -8 to +8, -8 represents the minimum tap position, and +8 represents the maximum tap position.

[0021] In one embodiment of the present invention, the step of preprocessing the first feature sequence includes: Step S201, using a box plot to remove outliers in the first feature sequence; Step S202, calculating the ratio of voltage to current, obtaining load change characteristics, and splicing them to the power data; Step S203, normalizing the first feature sequence by using a z-score normalization method to obtain a second feature sequence.

[0022] In one embodiment of the present invention, the voltage prediction model comprises: a hidden layer, an extraction layer and a classifier; The hidden layer includes N first units, the nth first unit is used to input the nth sequence unit of the second feature sequence, and output a feature vector; The extraction layer is used to extract the feature vector output by the Nth first unit as an update vector; The update vector is input into a classifier, and the classification space of the classifier represents the voltage in the second preset time period.

[0023] In one embodiment of the present invention, the hidden layer is constructed based on an improved gated recurrent unit, and the improvements include: introducing an adjustment coefficient in the update gate, introducing a scaling factor in the reset gate, introducing a residual connection, and adding a trend gate; the introduction of the adjustment coefficient can adjust the influence of the hidden state at the previous moment, allowing the model to respond more quickly to short-term fluctuations in voltage, thereby improving the flexibility and accuracy of the prediction; the introduction of the scaling factor can reduce the interference of the reset gate on short-term fluctuations, helping the model to more stably retain the long-term trend information of the voltage, avoiding excessive response to short-term noise, and improving the stability of the model; the introduction of the residual connection can retain the direct influence of the original input data, preventing the model from losing key information when processing short-term fluctuations, and enabling the details of the voltage fluctuations to be better reflected in the prediction results; adding a trend gate is used to retain the long-term change characteristics of the voltage, enhancing the model's learning ability for voltage changes over a long period of time, so that the model can simultaneously process short-term fluctuations and long-term trends, thereby improving the overall prediction accuracy.

[0024] Specifically, the calculation formula of the nth first unit of the hidden layer includes: The calculation formula of the update gate of the first unit of the hidden layer n is: ; in, represents the update gate of the first unit of the nth hidden layer, Represents the nth sequence unit of the second feature sequence currently input, represents the feature vector output by the first unit of the n-1th cell, , and They respectively represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the update gate of the first unit of the nth hidden layer, Represents the adjustment coefficient, which is used to adjust the hidden state of the previous moment The weight in the update gate increases the sensitivity of the update gate to short-term changes and enhances its responsiveness to rapid voltage changes; The calculation formula of the reset gate of the first unit of the hidden layer n is: ; in, represents the reset gate of the first unit of the nth hidden layer, , and They respectively represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the reset gate of the first unit of the nth hidden layer, represents the scaling factor, which is used to adjust the response speed of the reset gate to the current input and hidden state, prevent the network from overreacting to short-term noise, and retain long-term information; The calculation formula of the trend gate of the first unit of the nth hidden layer is: ; in, represents the trend gate of the first unit of the nth hidden layer, , and They respectively represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the trend gate of the first unit of the nth hidden layer; The calculation formula for the candidate hidden state of the nth first unit in the hidden layer is: ; in, represents the candidate hidden state of the first unit of the hidden layer n, , and Respectively represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the candidate hidden state of the nth first unit of the hidden layer; The calculation formula for updating the hidden state of the first unit of the hidden layer is: ; in, represents the feature vector output by the first unit of the hidden layer n, represents point-wise multiplication, Represents the weight coefficient of the residual connection, which controls the direct impact of the current input on the hidden state and prevents excessive smoothing of information or loss of details of rapid voltage changes.

[0025] In one embodiment of the present invention, a control scheme is generated based on a voltage prediction model combined with a genetic algorithm that introduces adaptive t-distribution disturbance variation, including the following steps: Step S301, generating an initialization population with Q individuals meeting the constraint conditions, where Q is a custom parameter, preferably, Q is set to 20; The constraints include: The voltage of new energy stations needs to be maintained within a preset safety range to avoid excessively high or low voltages from having adverse effects on equipment or system operation; When the reactive power compensation equipment is in operation, its output reactive power is limited by the preset capacity and cannot exceed the maximum output capacity of the equipment; The tap position of the voltage regulator can only be adjusted within a preset range; The individuals of the initialized population are represented by vector coding, which is represented as follows: ,in They represent the start and stop status of the first reactive power compensation device to the Jth reactive power compensation device respectively. Represented by binary code, 0 means the device is turned off, 1 means the device is turned on, They represent the positions of the 1st regulator tap to the Kth regulator tap, respectively. Represented by integer encoding; For example The value assigned as 0 indicates that the third reactive power compensation device is in the off state. The assigned value of +4 means that the tap position of the fourth voltage regulator is at +4; Step S302, constructing an objective function by combining a voltage prediction model; Step S303, obtaining the fitness value of each individual in the initialized population through objective function calculation; Step S304, selecting two individuals with the largest fitness values ​​in the initialization population as parents; Step S305, generating a random number in the range of 0 to 1, and when the random number is less than a first threshold, performing a crossover operation that satisfies the constraint condition; The first threshold is a custom parameter. Preferably, the first threshold is set to 0.7; The crossover operation means selecting the element value of one position of the vector encoding of two individuals in the parent generation to exchange; For example, an individual in the parent generation corresponds to a vector encoding The vector encoding corresponding to another individual in the parent generation Make an exchange; Step S306, mutating the individuals in the population based on the adaptive t-distribution perturbation mutation strategy; Step S307, if it is determined that the current number of iterations is greater than or equal to the maximum number of iterations, the individual with the largest fitness value in the initialized population is used as the control scheme, otherwise return to step S303, where the maximum number of iterations is a custom parameter, preferably, the maximum number of iterations is set to 200.

[0026] In one embodiment of the present invention, the objective function of step S302 is: ; in represents the objective function, represents the voltage output by the voltage prediction model, Indicates the preset target voltage, Represents a custom minimum positive value, Indicates the start and stop status of the jth reactive compensation device, It indicates the adjustment amplitude of the kth voltage regulator tap relative to the previous moment.

[0027] The objective function comprehensively considers minimizing voltage deviation, reducing frequent start and stop of reactive power compensation equipment and reducing large-scale adjustment of voltage regulator, and optimizes voltage regulation by finding a solution that maximizes the objective function.

[0028] In one embodiment of the present invention, the specific steps of step S306 include: Step S3061, initializing the degrees of freedom of the t distribution. Preferably, the degrees of freedom are set to 2, which can make the tail of the distribution thicker and enable a large-scale global search. Among them, the probability density function of t distribution is: ; Step S3062: dynamically adjust the degree of freedom according to the number of iterations. The degree of freedom adjustment formula is: ,in represents the current degrees of freedom, and denote the initial and final degrees of freedom, respectively. and P represent the current number of iterations and the preset maximum number of iterations respectively; It should be noted that in the early iteration stage, it is necessary to explore the global solution space, keep the degrees of freedom small, make the tail of the t distribution thicker, enhance the ability of large-step search, and explore potential areas far away from the current solution; in the middle and late iteration stages, in order to gradually converge, the degrees of freedom are appropriately increased, the t distribution is closer to the normal distribution, the mutation operation is concentrated near the solution, and local fine search is performed; Step S3063, mutate the current individual by applying t distribution to generate a new individual. The formula for generating the new individual is: ,in Represents a new individual, Represents the current individual, represents a t-distributed random number generated according to the value of the degrees of freedom, Represents the scaling factor that controls the strength of the mutation.

[0029] In one embodiment of the present invention, a smooth transition from global to local search is achieved by using a genetic algorithm that introduces an adaptive t-distribution perturbation mutation strategy. The adaptive adjustment of the degrees of freedom can flexibly adjust the amplitude and scope of the mutation according to the progress stage of the problem, thereby improving the optimization ability of the genetic algorithm in a complex search space and avoiding the problem of local optimal solutions.

[0030] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are within the protection of the present embodiment.

Claims

1. A voltage control and optimization system for new energy stations based on real-time monitoring, characterized in that: include: A data acquisition module is used to collect power data and equipment status data of a new energy station within a first preset time period T according to a preset time interval t, and represent it through a first characteristic sequence, wherein the nth sequence unit of the first characteristic sequence represents the power data and equipment status data at the nth time point, wherein T and t are custom parameters, N=T / t, 1≤n≤N; The power data includes: voltage, current, active power and reactive power; the equipment status data includes: the start and stop status of J reactive compensation devices and the position of K voltage regulator taps; A data preprocessing module, which is used to preprocess the first feature sequence to obtain a second feature sequence; A voltage prediction module, which is used to construct a voltage prediction model according to the second characteristic sequence, wherein the input of the voltage prediction model is the second characteristic sequence, and the output is the voltage in the second preset time period; A control scheme generating module, which is used to generate a control scheme according to the voltage in the second preset time period in combination with a genetic algorithm that introduces adaptive t-distribution disturbance variation; The start and stop states of J reactive compensation devices and the positions of K voltage regulator taps in the second preset time period are determined according to the control scheme.

2. According to the new energy station voltage control optimization system based on real-time monitoring according to claim 1, it is characterized in that: The start and stop states of the reactive power compensation equipment include: enabled and disabled, and the encoding method thereof adopts binary encoding; the position of the voltage regulator tap is a discrete value and adopts integer encoding.

3. According to the new energy station voltage control optimization system based on real-time monitoring according to claim 1, it is characterized in that: The step of preprocessing the first feature sequence comprises: Step S201, using a box plot to remove outliers in the first feature sequence; Step S202, calculating the ratio of voltage to current, obtaining load change characteristics, and splicing them to the power data; Step S203, normalizing the first feature sequence by using a z-score normalization method to obtain a second feature sequence.

4. According to the new energy station voltage control optimization system based on real-time monitoring according to claim 1, it is characterized in that: The voltage prediction model includes: a hidden layer, an extraction layer and a classifier; The hidden layer includes N first units, the nth first unit is used to input the nth sequence unit of the second feature sequence, and output a feature vector; The extraction layer is used to extract the feature vector output by the Nth first unit as an update vector; The update vector is input into a classifier, and the classification space of the classifier represents the voltage in the second preset time period.

5. A voltage control and optimization system for new energy stations based on real-time monitoring according to claim 4, characterized in that: The hidden layer is constructed based on an improved gated recurrent unit, and the improvements include: introducing an adjustment coefficient in an update gate, introducing a scaling factor in a reset gate, introducing a residual connection, and adding a trend gate.

6. The voltage control and optimization system for new energy stations based on real-time monitoring according to claim 1 is characterized in that: The control scheme is generated based on the voltage prediction model combined with the genetic algorithm that introduces adaptive t-distribution disturbance variation, including the following steps: Step S301, generating an initialization population with Q number of individuals meeting the constraint conditions, where Q is a custom parameter; Step S302, constructing an objective function by combining a voltage prediction model; Step S303, obtaining the fitness value of each individual in the initialized population through objective function calculation; Step S304, selecting two individuals with the largest fitness values ​​in the initialization population as parents; Step S305, generating a random number in the range of 0 to 1, and when the random number is less than a first threshold, performing a crossover operation that satisfies the constraint condition; The crossover operation means selecting the element value of one position of the vector encoding of two individuals in the parent generation to exchange; Step S306, mutating the individuals in the population based on the adaptive t-distribution perturbation mutation strategy; Step S307, if it is determined that the current number of iterations is greater than or equal to the maximum number of iterations, the individual with the largest fitness value in the initialized population is used as the control scheme, otherwise, return to step S303, where the maximum number of iterations is a custom parameter.

7. A voltage control and optimization system for new energy stations based on real-time monitoring according to claim 6, characterized in that: The objective function of step S302 is: ； in represents the objective function, represents the voltage output by the voltage prediction model, Indicates the preset target voltage, Represents a custom minimum positive value, Indicates the start and stop status of the jth reactive compensation device, It indicates the adjustment amplitude of the kth voltage regulator tap relative to the previous moment.

8. The voltage control and optimization system for new energy stations based on real-time monitoring according to claim 6 is characterized in that: The specific steps of step S306 include: Step S3061, initializing the degrees of freedom of the t distribution, wherein the probability density function of the t distribution is: ； Step S3062: dynamically adjust the degree of freedom according to the number of iterations. The degree of freedom adjustment formula is: ,in represents the current degrees of freedom, and denote the initial and final degrees of freedom, respectively. and P represent the current number of iterations and the preset maximum number of iterations respectively; Step S3063, mutate the current individual by applying t distribution to generate a new individual. The formula for generating the new individual is: ,in Represents a new individual, Represents the current individual, represents a t-distributed random number generated according to the value of the degrees of freedom, Represents the scaling factor that controls the strength of the mutation.

Images