Dynamic adaptive adjustment method and system based on wind power generation operation data

Abstract

The invention discloses a dynamic adaptive adjustment method and system based on wind power generation operation data, and the method comprises the steps: building a data driving model based on physical constraints through collecting the power grid voltage fluctuation and wind power output power change data of a wind power generation grid connection link in real time, and setting a reference model of a model reference adaptive control algorithm; and inputting real-time data into the model to obtain a wind power output voltage predicted value, comparing the predicted value with a reference model standard parameter to calculate a deviation, adjusting a control parameter accordingly, and further controlling a converter and a reactive power compensation device to dynamically adjust the wind power output voltage so as to enable the wind power output voltage to be matched with the power grid voltage. The system comprises a data real-time acquisition unit, a data processing modeling unit and other six mutually connected units which respectively execute the functions of data acquisition, model establishment, parameter setting and calculation, equipment control and the like, the wind power generation grid-connected stability and the electric energy quality are effectively improved, and the adaptability of the system to power grid conditions is enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wind power generation operation data management, and particularly relates to a dynamic adaptive adjustment method and system based on wind power generation operation data. BACKGROUND

[0002] With the increasing demand for clean energy worldwide, wind power generation, as a green and environmentally friendly renewable energy generation method, has an increasing share in the power supply field. However, the intermittent and fluctuating nature of wind power generation poses many challenges in the grid-connection process. In the process of wind power generation, unpredictable changes in wind speed can lead to unstable wind power output, while grid voltage fluctuations can be caused by various factors, which requires wind power generation systems to have the ability to quickly adapt to changes in grid conditions to ensure stable operation of the power system and power quality.

[0003] Currently, traditional wind power grid-connection adjustment techniques have certain limitations. On the one hand, existing adjustment methods mostly use fixed control strategies, which are difficult to adjust flexibly according to real-time changes in grid conditions and wind power output characteristics. When the grid voltage fluctuates greatly or the wind power output power changes dramatically, the fixed control strategy cannot effectively adjust the wind power output voltage in time, resulting in a decrease in the matching degree between wind power and the grid, and thus affecting the stability and power quality of the wind power generation system. On the other hand, some model-based adjustment methods do not fully consider actual operating parameters and physical constraints. When building models, these methods often ignore the changes in device parameters in the wind power generation system with environmental factors (such as temperature, humidity), as well as the nonlinear characteristics in the system operation process, which limits the prediction accuracy and adjustment effect of the model, making it difficult to meet the requirements of wind power grid-connection for adjustment accuracy and response speed. SUMMARY

[0004] In order to overcome the shortcomings and deficiencies of the prior art, the present application provides a dynamic adaptive adjustment method and system based on wind power generation operation data.

[0005] The technical solution adopted by the present application is a dynamic adaptive adjustment method based on wind power generation operation data, comprising the following steps:

[0006] Step S1: Real-time collection of grid voltage fluctuation in the wind power grid-connection process is performed, and data collection devices are used to obtain continuous change data of grid voltage in the time domain, and at the same time, change data of wind power output power at different time nodes is collected, and the collected grid voltage data and wind power output power data are integrated to form an initial data set;

[0007] Step S2: based on the initial data set, a data-driven model based on physical constraints is established, which takes the grid voltage fluctuation amplitude, frequency of change and the fluctuation range, trend of change of wind power output power as input variables, combines the physical characteristics of the equipment in the wind power system, and constructs the mapping relationship between the data;

[0008] Step S3: set the reference model in the model reference adaptive control algorithm, and match the operating parameters when the grid voltage and the wind power output voltage are matched under the ideal state as the standard parameters of the reference model, including reference voltage amplitude, reference voltage phase and reference power factor;

[0009] Step S4: input the real-time collected operating data into the established data-driven model based on physical constraints, and calculate the predicted value of the wind power output voltage under the current state through the model;

[0010] Step S5: compare the predicted value with the standard parameters in the reference model, calculate the deviation value between them, and adjust the control parameters by using the model reference adaptive control algorithm according to the deviation value;

[0011] Step S6: according to the adjusted control parameters, control the wind turbine generator set converter and reactive power compensation device to dynamically adjust the wind power output voltage, so that the wind power output voltage is matched with the grid voltage.

[0012] Further, the data-driven model based on physical constraints established in step S2 is expressed as:

[0013] V pred =f(P wind ,ΔV grid ,R s ,X s ,P rated ,V rated )

[0014] Wherein, V pred is the predicted value of wind power output voltage; P wind is the real-time collected wind power output power; ΔV grid is the grid voltage fluctuation; R s is the line resistance parameter of the wind power system; X s is the line reactance parameter of the wind power system; P rated is the rated power of the wind turbine generator set; V rated is the rated voltage of the wind turbine generator set; and function f represents the mapping relationship constructed based on physical constraints.

[0015] Further, the calculation formula of the reference voltage amplitude V ref-amp of the reference model in step S3 is:

[0016] V ref-amp =V grid-norm ×(1±δ)

[0017] Among them, V grid-norm δ represents the standard voltage amplitude during normal grid operation; δ is the set voltage amplitude fluctuation tolerance coefficient.

[0018] Furthermore, the formula for calculating the deviation value in step S5 is as follows:

[0019]

[0020] Where ∈ represents the comprehensive deviation value; V pred This is the predicted output voltage value of the wind power unit; V ref-amp Reference voltage amplitude; This is the predicted phase value of the wind power output voltage; This is the reference voltage phase.

[0021] Furthermore, in step S5, when adjusting the control parameters using the model reference adaptive control algorithm, the control parameter K... adj The adjustment formula is:

[0022]

[0023] Among them, K adj The adjusted control parameters; K init γ is the initial value of the control parameter; γ is the adaptive adjustment gain coefficient; ∈ is the calculated deviation value; Let J be the gradient of the objective function J with respect to the control parameter K. The objective function J is constructed based on the degree of matching between the wind power output voltage and the grid voltage.

[0024] Furthermore, in step S6, when controlling the wind turbine generator converter, the converter output voltage regulation amount ΔV conv The formula for calculation is:

[0025] ΔV conv =K conv ×(V ref-amp -V pred )

[0026] Where, ΔV conv K represents the output voltage regulation of the converter. conv V is the converter voltage regulation coefficient, which is determined by the converter's hardware characteristics and control strategy. ref-amp Reference voltage amplitude; V pred This is the predicted value of the wind power output voltage.

[0027] Furthermore, in step S6, when controlling the reactive power compensation device, the reactive power compensation amount Q... compThe calculation formula of V is:

[0028]

[0029] Wherein, Q comp is the reactive power compensation of the reactive power compensation device; K q is the reactive power compensation coefficient, which is determined according to the type and capacity of the reactive power compensation device and the reactive-voltage characteristic of the wind power generation system; is the reference voltage phase; is the wind power output voltage phase prediction value; P wind is the real-time collected wind power output power.

[0030] Further, in the step S4, when the real-time collected operation data is input into the data-driven model based on physical constraints, the model is modified in combination with the influence of the temperature T on the equipment parameters in the wind power generation system, and the modified model is:

[0031] V pred-mod = f(P wind , ΔV grid , R s (T), X s (T), P rated , V rated )

[0032] Wherein, V pred-mod is the modified wind power output voltage prediction value; R s (T) is the line resistance parameter combined with the influence of the temperature T; X s (T) is the line reactance parameter combined with the influence of the temperature T; and the function f constructs a new mapping relationship according to the physical influence law of the temperature on the equipment parameters.

[0033] Further, in the step S3, the reference power factor cos of the reference model is calculated according to the following formula:

[0034]

[0035] Wherein, P grid－base is the grid reference active power; and Q grid－base is the grid reference reactive power.

[0036] The dynamic self-adaptive adjustment system based on wind power generation operation data comprises:

[0037] A data real-time collection unit is configured to collect the grid voltage fluctuation data and the wind power output power change data in the wind power generation grid connection link in real time, and output the collected data.

[0038] The data processing modeling unit is connected with the data real-time acquisition unit, receives the acquired data, and establishes a data-driven model based on physical constraints.

[0039] The reference model setting unit is used for setting a reference model in a model reference adaptive control algorithm, and determining standard parameters of the reference model.

[0040] The prediction calculation unit is connected with the data processing modeling unit and the reference model setting unit, inputs the real-time acquisition data into the data-driven model to calculate a wind power output voltage prediction value, and compares the wind power output voltage prediction value with the standard parameters of the reference model to calculate a deviation value.

[0041] The parameter adjustment unit is connected with the prediction calculation unit, and adjusts the control parameters by using the model reference adaptive control algorithm according to the deviation value.

[0042] The device control unit is connected with the parameter adjustment unit, and controls the wind turbine generator set converter and the reactive power compensation device according to the adjusted control parameters to dynamically adjust the wind power output voltage.

[0043] Beneficial effects: The application proposes a dynamic adaptive adjustment method and system based on wind power generation operation data, which realizes accurate dynamic adjustment of the wind power output voltage by acquiring real-time grid voltage fluctuation and wind power output power change data, constructing a data-driven model based on physical constraints, and combining a model reference adaptive control algorithm. Unlike traditional fixed control strategies, this method can flexibly adjust control parameters according to real-time changes in grid conditions and wind power output characteristics, ensuring that the wind power output voltage and the grid voltage always remain well matched, greatly improving the stability and power quality of the wind power generation system connected to the grid. In view of the problem that the existing technology does not consider actual operation parameters and physical constraints in the model, the system fully considers the physical characteristics constraints of the equipment parameters in the wind power generation system when establishing the data-driven model, and also considers the influence of environmental factors such as temperature on the equipment parameters, comprehensively and accurately reflecting the system operation state. In actual operation, by continuously comparing the wind power output voltage prediction value with the standard parameters of the reference model, the deviation is quickly calculated and the control parameters are adaptively adjusted to drive the converter and the reactive power compensation device and other equipment for accurate adjustment, so that the system has stronger adaptability to complex grid conditions, effectively improves the model prediction accuracy and adjustment response speed, and meets the strict requirements of wind power grid connection on adjustment accuracy and real-time performance. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 The method steps flowchart of the application;

[0045] Figure 2 The system unit composition diagram of the application. DETAILED DESCRIPTION

[0046] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict, and the present application will be further described in detail below in combination with the drawings and specific embodiments.

[0047] As shown in the figure, the dynamic adaptive adjustment method based on wind power generation operation data includes the following steps: Figure 1

[0048] Step S1: Real-time collection of grid voltage fluctuation in wind power generation grid-connected link, using data acquisition device to obtain continuous change data of grid voltage in time domain, at the same time, collecting change data of wind power output at different time nodes, integrating the collected grid voltage data and wind power output data to form an initial data set;

[0049] Specifically, this step is to obtain the key data in the wind power generation grid-connected link in real time, and monitor the continuous change of grid voltage in time domain through professional data acquisition device. The time domain data here means that the acquisition device will record the numerical change of grid voltage continuously at certain time intervals, such as tens of times or even more per second, to ensure that the subtle fluctuation of voltage is captured. At the same time, the change data of wind power output at different time nodes is collected, which includes the real-time change value of output power of wind turbine during operation due to changes in wind speed, wind direction and other factors. After collecting the grid voltage data and wind power output data, they are integrated to form an initial data set. These data are the basis for all subsequent operations, and their accuracy and integrity directly affect the effect of the entire adjustment system.

[0050] From the implementation, the data acquisition device needs to have high precision and high reliability. Special designed sensors and data acquisition modules are usually used, which are installed at key nodes of wind power generation grid connection, such as the output end of wind turbine generator set, the access point of power grid and other positions. The sensor is responsible for converting physical quantities (voltage, power, etc.) into electrical signals, and the data acquisition module processes, amplifies, filters and other operations on the electrical signals, and then converts them into digital signals for subsequent storage and processing. When integrating data, the grid voltage data and wind power output data are corresponded one by one according to the unified time stamp standard, to ensure the correlation and time sequence of the data.

[0051] Step S2: Based on the initial data set, a data-driven model based on physical constraints is established, which takes the fluctuation amplitude, change frequency of grid voltage and the fluctuation range, change trend of wind power output as input variables, combines the physical characteristic constraints of equipment in wind power generation system, and constructs the mapping relationship between data;

[0052] ​Specifically, the purpose of this step is to build a model that can reflect the operation rules of the wind power generation system based on the initial data set. The model takes the grid voltage fluctuation amplitude, change frequency, and wind power output power fluctuation range and change trend as input variables. The grid voltage fluctuation amplitude refers to the degree of voltage deviation from its rated value within a certain time; the change frequency refers to the speed of voltage fluctuation. The wind power output power fluctuation range reflects the power variation interval, and the change trend reflects whether the power is rising, falling, or remaining stable. At the same time, the model combines the physical characteristics constraints of the equipment in the wind power generation system, such as the influence of line resistance and reactance on voltage, the rated power and rated voltage of the equipment on the operating range, etc., to build the mapping relationship between the data. In this way, the model can truly simulate the system's operating state under different operating conditions, providing a reliable basis for subsequent prediction and adjustment.

[0053] In the implementation process, the establishment of the model requires the use of professional data analysis and modeling techniques. First, the initial data set is analyzed in depth to mine the inherent rules and characteristics of the data. Statistical analysis, machine learning, etc. can be used to identify the potential relationship between the input variables and the wind power output voltage. Then, according to the physical principles and equipment characteristics of the wind power generation system, the relationship obtained by analysis is corrected and improved to ensure that the model meets the actual physical constraints. In the process of building the model, a large amount of testing and verification work needs to be done, and by comparing with the actual operation data, the parameters and structure of the model are constantly adjusted to improve the accuracy and reliability of the model.

[0054] Step S3: Set the reference model in the model reference adaptive control algorithm, and set the operating parameters when the grid voltage and wind power output voltage are matched in the ideal state as the standard parameters of the reference model, including reference voltage amplitude, reference voltage phase, and reference power factor;

[0055] Specifically, setting the reference model provides an ideal operating standard for the model reference adaptive control algorithm. In the process of wind power grid connection, when the grid voltage and wind power output voltage are well matched, the system is in an ideal operating state, and the operating parameters at this time are selected as the standard parameters of the reference model, including reference voltage amplitude, reference voltage phase, and reference power factor. The reference voltage amplitude determines the value size that the wind power output voltage should reach in the ideal case; the reference voltage phase ensures the consistency of the wind power output voltage and the grid voltage in phase, avoiding power loss and system instability caused by phase difference; the reference power factor reflects the reasonable allocation ratio of active power and reactive power in the ideal state. These standard parameters provide a clear target for subsequent comparison and adjustment.

[0056] In implementation, the setting of the reference model needs to consider multiple factors. On the one hand, according to the relevant standards and specifications of the power grid, the requirements for voltage amplitude, phase and power factor during normal operation of the power grid are determined. On the other hand, the characteristics of the wind power generation system itself, such as the rated parameters of the wind turbine generator, the performance of the converter and the reactive power compensation device, are also considered to determine reasonable reference values. In practical applications, the parameters of the reference model are not fixed, but can be adjusted according to changes in the grid conditions and the operating state of the wind power generation system to ensure that it always has reference value.

[0057] Step S4: input the real-time collected operation data into the established data-driven model based on physical constraints, and calculate the predicted value of the wind power output voltage under the current state through the model;

[0058] Specifically, this step inputs the real-time collected operation data into the already established data-driven model based on physical constraints. These real-time data contain the latest changes of the grid voltage and wind power output power at the current time, as well as other related parameters that may affect system operation. Through the data mapping relationship and calculation logic pre-constructed in the model, the input data are processed and analyzed, so as to calculate the predicted value of the wind power output voltage under the current state. This predicted value is based on the understanding of the system operation law by the model and the analysis of the current data, and it reflects the value that the wind power output voltage may reach under the current working condition. The accuracy of the predicted value is crucial for subsequent regulation operations, as it provides a basis for judging whether the system deviates from the ideal operating state.

[0059] In the implementation process, the input of data needs to be in accordance with the format and requirements specified by the model. Usually, the collected data are pre-processed, such as data cleaning, normalization, etc., to remove noise and outliers, so that the data meet the input standards of the model. Then, the processed data are input into the model using special algorithms and programs, and the calculation function of the model is called to calculate the predicted value. In order to ensure the accuracy and efficiency of the calculation, the calculation process of the model also needs to be optimized, using appropriate calculation methods and data structures to reduce the calculation time and resource consumption.

[0060] Step S5: compare the predicted value with the standard parameters in the reference model, calculate the deviation value between them, and adjust the control parameters using the model reference adaptive control algorithm according to the deviation value;

[0061] Specifically, the key of this step is to compare the calculated wind power output voltage prediction value with the standard parameters in the reference model. By calculating the deviation value between the two, the difference between the current system operation state and the ideal state can be clearly understood. The calculation of the deviation value covers the difference in voltage amplitude, phase and other aspects, fully reflecting the deviation of the system. According to this deviation value, the model reference adaptive control algorithm is used to adjust the control parameters. The adjustment of control parameters aims to make the system adjust towards the ideal operating state. By changing the value of the control parameters, the working state of the wind turbine converter, reactive power compensation device and other equipment can be affected, so as to realize the adjustment of wind power output voltage.

[0062] In implementation, comparison calculation and parameter adjustment need to use accurate algorithms and control strategies. First, a special comparison algorithm is used to compare the prediction value and the standard parameters parameter by parameter, calculate the deviation value of each parameter, and integrate these deviation values in a certain way to get an index that can represent the overall deviation. Then, according to this index, the adjustment rule in the model reference adaptive control algorithm is used to modify the control parameters accordingly. In the adjustment process, the dynamic characteristics and stability of the system need to be considered to avoid excessive adjustment leading to system oscillation or instability. At the same time, the change of deviation value needs to be monitored constantly, and the adjustment process needs to be optimized and corrected according to the actual situation.

[0063] Step S6: According to the adjusted control parameters, control the wind turbine converter, reactive power compensation device, and dynamically adjust the wind power output voltage to make the wind power output voltage match the grid voltage.

[0064] Specifically, this step controls the wind turbine converter, reactive power compensation device and other key equipment according to the adjusted control parameters. The converter, as one of the core devices for adjusting the wind power output voltage, can adjust the amplitude and phase of the output voltage by changing its control parameters, so that it is closer to the requirements of the grid voltage. The reactive power compensation device is mainly used to adjust the reactive power of the system, and by controlling its reactive power compensation amount, the quality of the wind power output voltage can be improved and the power factor of the system can be improved. Through the coordinated control of these devices, dynamic adjustment of the wind power output voltage is realized, so that the wind power output voltage matches the grid voltage, ensuring that the wind power generation system can be stably connected to the grid and improving the power quality and reliability of the system.

[0065] In the implementation process, the equipment control needs to rely on advanced control system and communication technology. The control system receives the adjusted control parameters and converts them into corresponding control signals, which are transmitted to the converter and reactive compensation device through the communication network. After receiving the control signals, these devices act according to the preset control logic to achieve the adjustment of the output voltage. In the control process, the running state of the equipment needs to be monitored in real time to ensure that the equipment can work as expected. At the same time, according to the actual adjustment effect, the control parameters are further fine-tuned to achieve the best adjustment effect.

[0066] Preferably, the data-driven model expression based on physical constraints established in step S2 is:

[0067]

[0068] wherein V pred is the wind power output voltage prediction value; P wind is the real-time collected wind power output; ΔV grid is the grid voltage fluctuation; R s is the line resistance parameter of the wind power generation system; X s is the line reactance parameter of the wind power generation system; P rated is the rated power of the wind turbine generator set; V rated is the rated voltage of the wind turbine generator set; and function f represents the mapping relationship constructed based on physical constraints, which is determined according to the physical characteristics of the equipment in the wind power generation system, such as the influence of line impedance on voltage, the limitation of equipment rated parameters, etc.

[0069] Specifically, the data-driven model based on physical constraints is constructed by taking the wind power output and the grid voltage fluctuation as input variables, combining the line resistance, the line reactance and other physical parameters of the equipment, and the rated power, the rated voltage and other equipment limitation conditions, to establish a mapping relationship that accurately reflects the correlation between the wind power output voltage and multiple factors. This model considers the physical characteristics constraints such as the influence of line impedance on voltage and the limitation of equipment rated parameters on operating range, and can accurately predict the wind power output voltage according to the real-time collected data, providing a reliable basis for subsequent control and adjustment. In the implementation process, the initial data set collected needs to be analyzed in depth to mine the internal laws in the data, and then professional data analysis and modeling techniques are used to construct a mapping relationship that conforms to the actual physical constraints, combined with the physical principles of the wind power generation system. Through a large number of testing and verification work, the model parameters and structure are continuously adjusted to improve the accuracy and reliability of the model prediction.

[0070] Preferably, the calculation formula of the reference voltage amplitude V ref-amp of the reference model in step S3 is:

[0071] V ref-amp = V grid-norm × (1 ± δ)

[0072] wherein, V grid-norm is the standard voltage amplitude when the power grid is in normal operation; δ is a set voltage amplitude fluctuation tolerance coefficient, which is determined according to the requirement of the power grid on voltage stability and the actual operation capacity of the wind power generation system, and is used to limit the reference voltage amplitude within a reasonable fluctuation range.

[0073] Specifically, the method for determining the reference voltage amplitude is based on the standard voltage amplitude when the power grid is in normal operation, and introduces a voltage amplitude fluctuation tolerance coefficient. The setting of the coefficient comprehensively considers the requirement of the power grid on voltage stability and the actual operation capacity of the wind power generation system, so that the reference voltage amplitude can fluctuate within a reasonable range. This design not only ensures the matching degree of the wind power output voltage and the grid voltage, but also provides certain adjustment flexibility for the system. In implementation, the requirement of the power grid on voltage amplitude when in normal operation needs to be determined according to the relevant standards and specifications of the power grid, and the reasonable tolerance coefficient value needs to be determined in combination with the characteristics of the wind power generation system itself, such as the rated parameters of the wind turbine generator set, the adjustment capacity of the converter, etc. The reference voltage amplitude will be adjusted appropriately according to the change of the power grid condition and the operation state of the wind power generation system, so as to ensure that it always has reference value.

[0074] Preferably, the formula for calculating the deviation value in the step S5 is:

[0075]

[0076] wherein, ∈ is the comprehensive deviation value; V pred is the predicted value of the wind power output voltage; V ref-a is the reference voltage amplitude; is the predicted value of the phase of the wind power output voltage; is the reference voltage phase, and the formula obtains a deviation value which comprehensively reflects the difference degree between the current predicted value and the reference value by calculating the square root of the sum of the square of the voltage amplitude deviation and the square of the phase deviation.

[0077] Specifically, the method for calculating the comprehensive deviation value is to calculate the voltage amplitude deviation and the phase deviation respectively, and then take the square root of the sum of the squares of the two to obtain an index that can comprehensively reflect the difference between the current wind power output voltage prediction value and the reference value. This index not only considers the difference in voltage amplitude, but also takes into account the influence of phase difference, and can more accurately quantify the deviation state of the system. In the implementation process, a special comparison algorithm is used to compare the prediction value and the reference value parameter by parameter, and the voltage amplitude and phase deviation values are calculated respectively. Then, according to a specific mathematical method, the two deviation values are comprehensively processed to obtain a comprehensive deviation value. This comprehensive deviation value provides an accurate quantitative basis for the subsequent adjustment of the control parameters, so that the system can be adjusted according to the actual deviation degree.

[0078] Preferably, when the model reference adaptive control algorithm is used to adjust the control parameters in step S5, the adjustment formula of the control parameter K adj is as follows:

[0079]

[0080] wherein K adj is the adjusted control parameter; K init is the initial value of the control parameter; γ is the adaptive adjustment gain coefficient, which is determined according to the dynamic response characteristics and stability requirements of the wind power generation system; ∈ is the calculated deviation value; and is the gradient of the objective function J with respect to the control parameter K, and the objective function J is constructed according to the matching degree of the wind power output voltage and the grid voltage.

[0081] Specifically, the control parameter adjustment formula is based on the initial control parameter value, combined with the adaptive adjustment gain coefficient, the calculated deviation value, and the gradient of the objective function with respect to the control parameter, to realize the dynamic adjustment of the control parameter. The determination of the adaptive adjustment gain coefficient takes into account the dynamic response characteristics and stability requirements of the wind power generation system, ensuring that the adjustment process is both fast and stable. The objective function is constructed according to the matching degree of the wind power output voltage and the grid voltage, reflecting the adjustment goal of the system. In implementation, first, the initial value of the control parameter is determined according to the historical operation data and characteristic analysis of the system. Then, during the operation of the system, the gradient of the deviation value and the objective function is calculated in real time, and the control parameter is updated according to the adjustment formula. By continuously monitoring the change of the deviation value, the adjustment process is optimized and corrected, ensuring that the system can quickly and stably approach the ideal operating state.

[0082] Preferably, when the wind turbine generator set converter is controlled in step S6, the calculation formula of the converter output voltage adjustment amount ΔV conv is as follows:

[0083] ΔV conv = Kconv X(V ref-amp -V pred )

[0084] wherein, AV conv is the output voltage adjustment of the converter; K conv is the voltage adjustment coefficient of the converter, which is determined by the hardware characteristics and control strategy of the converter; V re is the reference voltage amplitude; V pred is the predicted value of the wind power output voltage, and the adjustment amount of the output voltage of the converter is calculated through the formula to realize the matching between the wind power output voltage and the grid voltage.

[0085] Specifically, the calculation method of the output voltage adjustment of the converter is to accurately determine the voltage adjustment amount that needs to be output by the converter through the product of the voltage adjustment coefficient of the converter and the difference between the reference voltage amplitude and the predicted voltage value. The voltage adjustment coefficient of the converter is determined by the hardware characteristics and control strategy of the converter, which reflects the ability and characteristics of the converter to adjust the voltage. In the implementation process, the control system first obtains the reference voltage amplitude and the predicted wind power output voltage value, and calculates the difference between the two. Then, according to the specific model and performance parameters of the converter, the corresponding voltage adjustment coefficient is determined. The difference is multiplied by the adjustment coefficient to obtain the voltage adjustment amount that needs to be output by the converter. By accurately controlling the output voltage of the converter, the wind power output voltage amplitude is effectively adjusted to be closer to the requirement of the grid voltage.

[0086] Preferably, when the reactive power compensation device is controlled in the step S6, the calculation formula of the reactive power compensation amount Q comp is:

[0087]

[0088] wherein, Q comp is the reactive power compensation amount of the reactive power compensation device; K q is the reactive power compensation coefficient, which is determined according to the type and capacity of the reactive power compensation device and the reactive power-voltage characteristics of the wind power generation system; is the reference voltage phase; is the predicted value of the wind power output voltage phase; P wind is the real-time collected wind power output, and the required reactive power compensation amount is calculated through the formula to adjust the phase of the wind power output voltage.

[0089] Specifically, the calculation formula of the reactive power compensation quantity comprehensively considers the difference between the reference voltage phase and the predicted voltage phase and the real-time wind power output, and quantitatively calculates through a reactive power compensation coefficient. The reactive power compensation coefficient is determined according to the type and capacity of the reactive power compensation device and the reactive-voltage characteristics of the wind power generation system, so as to ensure that the calculated reactive power compensation quantity can effectively adjust the phase of the wind power output voltage. In implementation, the system first obtains the reference voltage phase and the predicted wind power output voltage phase, and calculates the difference between the two. At the same time, the wind power output power data is collected in real time. Then, according to the specific parameters of the reactive power compensation device and the reactive-voltage characteristics of the system, the reactive power compensation coefficient is determined. The phase difference, the wind power output power and the reactive power compensation coefficient are multiplied to obtain the required reactive power compensation quantity. By controlling the reactive power compensation device to output the corresponding reactive power, the adjustment of the wind power output voltage phase is realized, and the power factor and power quality of the system are improved.

[0090] Preferably, in the step S4, when the real-time collected operation data is input into the data-driven model based on physical constraints, the model is modified by combining the influence of the temperature T on the equipment parameters in the wind power generation system, and the modified model is:

[0091] V pred-mod = f (P wind , ΔV grid , R s (T), X s (T), P rated , V rated )

[0092] wherein V pred-mod is the modified wind power output voltage prediction value; R s (T) is the line resistance parameter combined with the influence of the temperature T; X s (T) is the line reactance parameter combined with the influence of the temperature T; the meanings of other parameters are the same as in claim 2, and the function f constructs a new mapping relationship according to the physical influence law of the temperature on the equipment parameters.

[0093] Specifically, in combination with the temperature influence model correction method, the temperature factor is considered on the basis of the original data-driven model, and the line resistance and reactance parameters are temperature corrected. The change of temperature will affect the physical properties of the equipment, and then affect the prediction accuracy of the wind power output voltage. By establishing the physical relationship model between temperature and equipment parameters, the original model is corrected, which can improve the prediction accuracy of the model under different environmental conditions. In the implementation process, first, the change data of equipment parameters under different temperature conditions need to be collected, and the influence law of temperature on line resistance and reactance is analyzed. Then, according to these laws, the temperature correction model is established, and the temperature is taken as a new input variable into the original data-driven model. In actual operation, real-time acquisition of environmental temperature data is carried out, and the line resistance and reactance parameters are adjusted in real time through the temperature correction model, so as to improve the prediction accuracy of the model for wind power output voltage, and make the system better adapt to different environmental conditions.

[0094] Preferably, in the step S3, the reference power factor cos of the reference model is calculated according to the following formula:

[0095]

[0096] wherein, P grid－base is the grid reference active power; Q grid－base is the grid reference reactive power, and the formula calculates the reference power factor in the ideal state according to the grid reference active power and reactive power, which is used as one of the reference standards of the model reference adaptive control algorithm.

[0097] Specifically, the reference power factor is determined by calculating the ratio of the grid reference active power and reactive power. This method determines the reference power factor in the ideal state based on the actual operating state of the grid, and provides an important reference standard for the model reference adaptive control algorithm. Accurate setting of the reference power factor helps to optimize the power quality of wind power output and improve the operating efficiency of the system. In implementation, first, the grid reference active power and reactive power data need to be obtained, which can usually be obtained from the grid dispatching center or related monitoring equipment. Then, according to the specific mathematical formula, the ratio of the two powers is calculated to obtain the reference power factor. The reference power factor is used as an important reference standard of the model reference adaptive control algorithm, which is compared with the power factor in actual operation, so as to adjust the control parameters and make the power factor of wind power output close to the ideal state, improve the power quality and system efficiency.

[0098] As shown in Figure 2 , the dynamic adaptive adjustment system based on wind power generation operation data includes:

[0099] The data real-time acquisition unit is used for acquiring grid voltage fluctuation data and wind power output power change data in real time in the wind power generation grid connection link and outputting the acquired data.

[0100] The data processing modeling unit is connected with the data real-time acquisition unit, receives the acquired data, and establishes a data-driven model based on physical constraints.

[0101] The reference model setting unit is used for setting a reference model in a model reference adaptive control algorithm and determining standard parameters of the reference model.

[0102] The prediction calculation unit is connected with the data processing modeling unit and the reference model setting unit, inputs real-time acquisition data into the data-driven model to calculate a wind power output voltage prediction value, and compares the wind power output voltage prediction value with the standard parameters of the reference model to calculate a deviation value.

[0103] The parameter adjustment unit is connected with the prediction calculation unit, adjusts control parameters according to the deviation value by using the model reference adaptive control algorithm.

[0104] The equipment control unit is connected with the parameter adjustment unit, controls a wind turbine generator set converter and a reactive power compensation device according to the adjusted control parameters to dynamically adjust wind power output voltage.

[0105] The dynamic adaptive adjustment method and system based on wind power generation operation data can accurately capture the dynamic characteristics of the wind power generation system by acquiring grid voltage fluctuation and wind power output power change data in real time, constructing a data-driven model based on physical constraints, and adopting a model reference adaptive control algorithm to automatically adjust control parameters according to real-time operation data, so that the wind power output voltage and the grid voltage can always be well matched, and the grid stability and power quality can be greatly improved.

[0106] The present application has unique advantages in overcoming the problem of insufficient consideration of physical constraints in traditional technologies. The data-driven model established by the present application fully considers the physical characteristic constraints of equipment in the wind power generation system, such as the influence of line impedance on voltage, the limitation of equipment rated parameters, and further considers the influence of environmental factors such as temperature on equipment parameters, so that the model prediction is more close to the actual operation condition. This comprehensive consideration of physical constraints in the modeling method effectively avoids the prediction deviation caused by the traditional model due to the neglect of key factors, improves the adaptability of the system to complex working conditions, and ensures accurate adjustment under various environmental conditions.

[0107] The adaptive adjustment mechanism of the present application is another highlight. By continuously comparing the wind power output voltage prediction value with the reference model standard parameter, the system can quickly calculate the deviation and dynamically adjust the control parameters to drive the converter and reactive power compensation device to make precise adjustment. This closed-loop control method not only has fast response speed, but also has strong robustness, and can effectively deal with complex situations such as large fluctuations in grid voltage and dramatic changes in wind power output. Compared with traditional technologies, the system has significant improvement in adjustment accuracy, response speed and stability, and can better meet the strict requirements of modern power systems for wind power grid connection, and promote the development of wind power technology to be more efficient and reliable.

[0108] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "arrangement", "installation", "connection", "connection", "fixing" should be understood broadly, for example, it can be fixed connection, or detachable connection, or integrally connected; it can be mechanical connection, or electrical connection; it can be directly connected, or indirectly connected through intermediate medium, or the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0109] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various equivalent changes, modifications, replacements and variations of the embodiments can be made without departing from the principles and spirits of the present application, and the scope of the present application is defined by the appended claims and their equivalent scope.

Claims

1. A dynamic self-adaptive adjustment method based on wind power generation operation data, characterized in that, Comprise the following steps: Step S1: Real-time acquisition of grid voltage fluctuation in wind power grid connection link, use data acquisition device to obtain continuous change data of grid voltage in time domain, at the same time, collect the change data of wind power output power at different time nodes, integrate the collected grid voltage data and wind power output power data to form an initial data set; Step S2: Based on the initial data set, a data-driven model based on physical constraints is established, which takes the fluctuation amplitude, frequency of grid voltage and the fluctuation range, trend of wind power output power as input variables, combines the physical characteristics constraints of equipment in wind power generation system, and constructs the mapping relationship between data; Step S3: Set the reference model in model reference adaptive control algorithm, and take the operating parameters when the grid voltage and wind power output voltage are matched in ideal state as the standard parameters of reference model, including reference voltage amplitude, reference voltage phase and reference power factor; Step S4: Input the real-time collected operating data into the established data-driven model based on physical constraints, and calculate the prediction value of wind power output voltage in current state through the model; Step S5: Compare the prediction value with the standard parameters in reference model, calculate the deviation value between them, and adjust the control parameters by using model reference adaptive control algorithm according to the deviation value; Step S6: According to the adjusted control parameters, control the wind turbine generator set converter and reactive power compensation device to dynamically adjust the wind power output voltage, so that the wind power output voltage and the grid voltage are matched.

2. The method of dynamic adaptive regulation based on wind power generation operation data according to claim 1, characterized in that, The expression of the data-driven model based on physical constraints established in step S2 is: V pred = f(P wind , ΔV grid , R s , X s , P rated , V rated ) Wherein, V pred is the wind power output voltage prediction value; P wind is the real-time collected wind power output; ΔV grid is the grid voltage fluctuation; R s is the wind power generation system line resistance parameter; X s is the wind power generation system line reactance parameter; P rated is the wind turbine rated power; V rated is the wind turbine rated voltage; and the function f represents the mapping relationship constructed based on physical constraints.

3. The method of claim 1, wherein, The reference voltage amplitude V of the reference model in the step S3 ref-amp The calculation formula is: V ref-amp = V grid-norm × (1 ± δ) Wherein, V grid-norm is the standard voltage amplitude when the power grid is in normal operation; δ is the set voltage amplitude fluctuation tolerance coefficient.

4. The method of claim 1, wherein, The formula for calculating the deviation value in step S5 is: Wherein, ∈ is the comprehensive deviation value; V pred is the wind power output voltage prediction value; V re is the reference voltage amplitude; is the wind power output voltage phase prediction value; is the reference voltage phase.

5. The method of claim 1, wherein, The adjustment formula of the control parameter K adj in the step S5 is: wherein K adj is the adjusted control parameter; K init is the initial value of the control parameter; γ is the adaptive adjustment gain coefficient; and ∈ is the calculated deviation value. is the gradient of the target function J with respect to the control parameter K, and the target function J is constructed according to the matching degree of the wind power output voltage and the grid voltage.

6. The method of dynamic adaptive regulation based on wind power generation operation data according to claim 1, characterized in that, In the step S6 of controlling the wind turbine generator set converter, the calculation formula of the converter output voltage adjustment amount AV conv is as follows. ΔV conv = K conv × (V ref-amp - V pred ) Wherein, ΔV conv is the output voltage regulation of the converter; K conv is the voltage regulation coefficient of the converter, which is determined by the hardware characteristics and control strategy of the converter; V ref-amp is the reference voltage amplitude; V pred is the predicted value of the output voltage of the wind power.

7. The method of dynamic adaptive regulation based on wind power generation operation data according to claim 1, characterized in that, The reactive power compensation amount Q comp is calculated by the following formula. wherein Q comp is a reactive power compensation amount of the reactive power compensation device; K q is a reactive power compensation coefficient, which is determined according to the type and capacity of the reactive power compensation device and the reactive power-voltage characteristic of the wind power generation system; is a reference voltage phase; is a wind power output voltage phase prediction value; P wind is a real-time collected wind power output power.

8. The method of dynamic adaptive regulation based on wind power generation operation data according to claim 2, characterized in that, In step S4, when the real-time collected operating data is input into the data-driven model based on physical constraints, the model is modified in combination with the influence of temperature T on equipment parameters in wind power generation system, and the modified model is: V pred-mod = f(P wind , ΔV grid , R s (T), X s (T), P rated , V rated ) Wherein, V pred-mod is the modified wind power output voltage prediction value; R s (T) is the line resistance parameter affected by the temperature T; X s (T) is the line reactance parameter affected by the temperature T; the function f constructs a new mapping relationship according to the physical influence law of temperature on the equipment parameter.

9. The method for dynamic adaptive regulation based on wind power generation operation data according to claim 1, characterized in that, In said step S3, the reference power factor of the reference model is calculated according to the formula: where P grid－base is the grid reference active power; Q grid－base is the grid reference reactive power.

10. A dynamic adaptive regulation system based on wind power generation operational data, characterized in that, Comprise: Data real-time acquisition unit, for real-time acquisition of grid voltage fluctuation data and wind power output power change data in wind power grid connection link, and output the collected data; Data processing modeling unit, connected with the data real-time acquisition unit, receives the collected data, and establishes a data-driven model based on physical constraints; Reference model setting unit, for setting the reference model in model reference adaptive control algorithm, and determining the standard parameters of reference model; Prediction calculation unit, connected with the data processing modeling unit and reference model setting unit, inputs real-time collected data into data-driven model to calculate wind power output voltage prediction value, and compares with reference model standard parameters to calculate deviation value; Parameter adjustment unit, connected with the prediction calculation unit, adjusts the control parameters by using model reference adaptive control algorithm according to the deviation value; Equipment control unit, connected with the parameter adjustment unit, controls the wind turbine generator set converter and reactive power compensation device according to the adjusted control parameters to dynamically adjust the wind power output voltage.

Images