Multi-direct-current participated frequency modulation coordination optimization method and system for hydroelectric power generation system

Abstract

The invention provides a multi-direct-current-participated frequency modulation coordination optimization method and system for a hydroelectric power generation system. The invention provides a unified state space modeling framework. A water turbine speed regulation system, an HVDC power modulation system, an AGC control system and the like are integrated in a unified model. Aiming at the problems of response redundancy, non-uniform distribution and insufficient combination with water and electricity in the existing multi-circuit extra-high voltage direct current line frequency modulation control, cross-resource collaborative optimization is carried out in combination with the water and electricity frequency modulation capability on the basis of considering the water energy efficiency; the invention provides a multi-channel HVDC-hydropower coordinated frequency modulation power coordination optimization control framework based on a multi-objective optimization algorithm, and dynamic linkage response between multiple HVDC and hydropower is realized by constructing a power regulation priority mechanism and a joint allocation algorithm. According to the invention, through unified modeling, parameter identification and multi-objective optimization strategies, the system frequency control performance and the frequency modulation resource configuration efficiency are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of frequency modulation control of hydroelectric systems, in particular to a multi-direct current participating hydroelectric system frequency modulation coordination optimization method and system. BACKGROUND

[0002] As a clean, efficient and fast-responding power source, hydropower has long been responsible for primary frequency modulation and part of secondary frequency modulation in China's power system, especially in water-rich regions such as Southwest China and Central China. The frequency modulation capacity of hydropower is the key to supporting system operation safety. However, with the continuous advancement of the "West-East Electricity Transmission" project, a large amount of hydropower in the southwest region is transmitted through multiple ultra-high voltage direct current lines, and the coupling relationship between the hydropower and the main grid has changed significantly. The timeliness and controllability of hydropower frequency modulation resources face new challenges. In particular, in an AC-DC hybrid system, when the power grid is disturbed or subjected to load fluctuations, the frequency stability problem is more prominent, and higher requirements are placed on the coordinated frequency modulation capacity between hydropower and direct current systems.

[0003] Currently, the frequency modulation resources used to support frequency stability in the power grid mainly include hydropower, thermal power, energy storage and direct current transmission systems. Among them, thermal power frequency modulation has certain inertia and response capacity, but is subject to climbing constraints and coal consumption economy, and has been difficult to meet the demand for rapid frequency modulation; energy storage frequency modulation has fast response and high precision, but is high in cost and limited in scale; direct current transmission systems have been gradually integrated into the frequency regulation system in recent years, especially through frequency power modulation (FPC) means, which can complete large power active regulation within tens of milliseconds and has great potential for rapid support. The hydro-turbine-governor system, as the oldest frequency modulation means, still has the significant advantages of stable operation, sufficient capacity and good economy.

[0004] However, the existing technology still has many problems in frequency modulation models, control strategies and optimized scheduling, including the following points: First, the modeling of various frequency modulation resources is independent, making it difficult to analyze and optimize in coordination. Currently, in actual engineering and dispatching practice, the hydro-turbine-governor system, HVDC (high voltage direct current transmission) power modulation device and AGC (automatic generation control) control system usually belong to different control levels and are developed based on different modeling assumptions and time domain characteristics. Their parameter structures, control objectives and response paths are different, making it difficult to form a unified framework in multi-resource coordinated modeling and optimized control. This model heterogeneity limits the deep coordination between resources and cannot support rapid parameter updating and integrated dispatching control in large systems.

[0005] Second, the water and electricity frequency modulation performance is significantly constrained by water head, water level, etc., and it is difficult to balance response speed and economy. The traditional setting method of the water and electricity governor mostly relies on experience parameters or simple PID optimization, and the target function is mainly frequency deviation, without fully considering the complex influencing factors such as water head loss, water energy utilization efficiency and guide vane response constraint of the hydraulic system, resulting in problems such as "frequency modulation is excessive, economy is insufficient" or "response is too strong, water head fluctuates greatly" in actual operation.

[0006] Third, the frequency modulation strategy of multiple HVDC channels lacks coordination, which is easy to cause uneven resource allocation. With the operation of multiple ±800kV and ±1100kV HVDC lines in the power grid, the power regulation capacity is crucial to frequency support. However, most of the current HVDC frequency modulation strategies use fixed proportion regulation or simple power-frequency linear mapping, without considering the mutual influence between multiple channels or coordinating and optimizing with local resources such as hydropower, which is easy to cause excessive regulation of some HVDC lines, resource waste and stability decline.

[0007] Fourth, the optimization target is single, and the frequency stability, economy and system constraints are not comprehensively considered. Most of the existing optimization methods only consider minimizing the frequency deviation (such as IACE or ITAE), and pay insufficient attention to other key performance indicators such as water energy utilization rate and HVDC regulation capacity occupancy rate, resulting in poor optimization results at the operation level, such as "frequency is stable, water is wasted, and HVDC is full", which fails to balance the frequency modulation effect and power transmission efficiency.

[0008] Therefore, in order to realize the dual goals of power system frequency stability and resource collaborative utilization, a coordinated control strategy is needed for AC / DC hybrid systems, which integrates multi-frequency modulation resource modeling and optimization, can meet the system constraint conditions, and realize dynamic coordinated frequency modulation of hydropower and HVDC channels through model unification, parameter identification and multi-objective optimization, so as to improve the overall frequency modulation efficiency and operation stability of the power grid. SUMMARY

[0009] The present application aims to at least solve one of the technical problems existing in the prior art.

[0010] To this end, the first aspect of the present application provides a hydropower system frequency modulation coordination optimization method with multiple HVDC participations.

[0011] The second aspect of the present application provides a hydropower system frequency modulation coordination optimization system with multiple HVDC participations.

[0012] The present application provides a hydropower system frequency modulation coordination optimization method with multiple HVDC participations, which comprises: A unified state space model of frequency modulation coordination optimization of a hydropower system is established, and the unified state space model comprises a hydraulic turbine speed regulation system, a synchronous generator, a frequency modulation system of a high-voltage direct-current (HVDC) converter station and an AGC control system. Based on real-time operation data and historical monitoring data, parameters in the unified state space model are jointly identified by using a fusion method of a least square method and an evolutionary algorithm. Based on the unified state space model and the parameter identification result, a comprehensive optimization objective function considering frequency deviation, water energy utilization efficiency and HVDC regulation capacity occupation is constructed, and optimal control parameters are output. Based on a rolling optimization mechanism, the control parameters are sent to a hydropower speed regulator, an HVDC regulation device and an AGC execution system, and dynamic collaborative scheduling of various frequency modulation resources is realized.

[0013] According to the hydropower system frequency modulation coordination optimization method participated by multiple direct currents in the technical solution of the present application, the following additional technical features can be further provided. In the above technical solution, the state space structure of the unified state space model is represented as:

[0014] In the above technical solution, the state space structure of the unified state space model is represented as:

[0015] In the above technical solution, the parameters in the unified state space model are jointly identified by using the fusion method of the least square method and the evolutionary algorithm, and the joint identification comprises: Input data is acquired, and the input data comprises wide area measurement system data during a frequency modulation event, EMS historical data, substation SCADA data and PMU synchronous phasor data; An identification model is constructed, and the identification model extracted based on the unified state space model is represented as:

[0016] In the above technical solution, the state space structure of the unified state space model is represented as: represents a parameter vector to be identified; represents a joint differential structure composed of various modeling equations; A unit parameter manual or a historical setting value is used as an initial reference to set an initial value of the identification parameter; The least square method and the evolutionary algorithm are combined to optimize the parameters, wherein the least square method is used to identify the linear time delay characteristics of the linear subsystem, and the evolutionary algorithm is used for global search of the nonlinear subsystem; the target function of the global search is:

[0017] wherein, represents the model output; represents the actual measured value; N represents the number of sampling points; The local hill climbing method is used to refine the local minimum value, and the disturbance cooperative evolution is introduced to avoid falling into the local optimum until the algorithm converges.

[0018] In the above technical solution, the least square method adopts an OLS algorithm or an RLS algorithm; And / or, the evolutionary algorithm adopts a differential evolution algorithm or a particle swarm optimization algorithm.

[0019] In the above technical solution, when the parameter identification is performed, a segmented time window mode is adopted to gradually approach the dynamic segment.

[0020] In the above technical solution, based on the unified state space model and the parameter identification result, a comprehensive optimization target function considering the frequency deviation, the water energy utilization efficiency and the HVDC regulation capacity occupation is constructed, and optimal control parameters are output, including: The performance indicators are constructed from three dimensions of system operation performance, energy efficiency and regulation capacity, including a frequency response performance indicator, a water energy utilization efficiency indicator and an HVDC regulation capacity occupation indicator; The comprehensive optimization target function is constructed based on the three performance indicators as follows:

[0021] wherein, represents the frequency response performance indicator; represents the water energy utilization efficiency indicator; represents the HVDC regulation capacity occupation indicator; , , All are adjustable weight parameters; The optimization variables include the turbine governor parameters, the HVDC regulation slope and the AGC region active frequency modulation distribution coefficient; The constraint conditions include the frequency constraint, the water level / water head constraint, the HVDC regulation power constraint and the dynamic response constraint; A multi-objective search method based on the evolutionary strategy is adopted to construct a parallel search framework, and based on the constructed search framework, the optimal control parameters are searched under the constraint conditions with the comprehensive optimization target function as the target.

[0022] In the technical scheme, the frequency response performance index is used to measure the recovery ability of the system frequency deviation after the disturbance response, and is expressed as:

[0023] wherein, represents the regional ACE signal; and T represents a time period. The water energy utilization efficiency index is used to measure the water energy waste or non-optimal operation loss caused by frequent regulation, and is estimated based on an empirical model constructed based on the water level fluctuation and power generation efficiency degradation curve. The HVDC regulation capacity occupation index is used to reflect the average proportion of the HVDC channel regulation power amplitude caused by frequency modulation, and is expressed as:

[0024] wherein, represents the change amount of the HVDC channel regulation power at time t; and represents the maximum regulation power amplitude of the HVDC channel.

[0025] In the technical scheme, the multi-objective search method adopts the NSGA-III algorithm, and the search process includes: initializing a plurality of solution vector groups; performing cross and mutation operations under the target function and constraint; introducing a Pareto sorting and crowding distance selection mechanism; constantly evolving to generate a new solution group, and gradually approaching the Pareto optimal frontier.

[0026] In the technical scheme, the control parameters are issued to the hydroelectric governor, the HVDC regulation device and the AGC execution system based on the rolling optimization mechanism, and the control parameters include: The control parameters are deployed to each control terminal of the system in a periodic updating manner, wherein the control parameters are periodically called in the energy management system of the power grid dispatching master station, and the controller parameters are dynamically adjusted in combination with the latest load prediction information, unit state data and disturbance detection results.

[0027] The application provides a multi-direct-current participated hydroelectric power system frequency modulation coordination optimization system, which includes: A unified modeling module is configured to establish a unified state space model of the hydroelectric power system frequency modulation coordination optimization, and the unified state space model includes a hydro-turbine governing system, a synchronous generator, a high-voltage direct-current power transmission converter station frequency modulation system and an AGC control system. A parameter identification module is configured to jointly identify parameters in the unified state space model based on real-time operation data and historical monitoring data by using a least square method and an evolutionary algorithm fusion method. A multi-objective optimization module, based on a unified state space model and parameter identification results, constructs a comprehensive optimization objective function considering frequency deviation, water energy utilization efficiency and HVDC regulation capacity occupation, and outputs optimal control parameters; A control deployment module, based on a rolling optimization mechanism, issues control parameters to hydroelectric governor, HVDC regulation device and AGC execution system, and realizes dynamic collaborative scheduling of various frequency modulation resources.

[0028] In summary, due to the adoption of the above technical features, the present application has the following advantages: The present application constructs a multi-source frequency modulation coordination control method suitable for AC-DC hybrid systems, breaks through the limitations of existing systems in resource modeling heterogeneity, control response fragmentation and single optimization target, and forms a unified regulation framework for new power system frequency stability and resource efficient collaboration. The method improves the dynamic response capability of hydroelectric and DC systems under frequency disturbance through integrated modeling, control strategy innovation and multi-objective optimization design, and realizes dual improvement of frequency modulation quality and operation benefit.

[0029] Specifically, in the background of current multi-type frequency modulation devices (such as hydroelectric generators and ultra-high voltage DC) constructing dynamic models and setting parameters independently, the present application proposes a modular modeling mechanism, which uniformly represents various active control resources based on a unified state space structure. The general modeling framework is introduced to express the dynamic behavior of water power, DC, AGC and other resources in a unified form, realizing parameter normalization and identification collaboration of control objects. Thus, on the basis of compatibility modeling, system-level joint identification and collaborative control strategy design are supported, reducing response conflicts and resource waste caused by traditional controller isolated design.

[0030] In view of the problem of repeated adjustment strategy and redundant response of multiple HVDC lines, the present application proposes a collaborative optimization method of multiple HVDC lines and hydroelectric power response based on system disturbance dynamic analysis. Based on the unified modeling and parameter identification module, the frequency deviation response, water energy consumption and HVDC regulation capacity are considered comprehensively, and the optimal frequency modulation control parameters are output through the constraint optimization method, finally realizing the performance optimization of hydroelectric-multiple DC joint frequency modulation. According to the adjustment boundary, dynamic margin and frequency contribution capacity of each channel, hierarchical scheduling is realized to achieve cross-channel power collaborative regulation.

[0031] Traditional frequency modulation optimization mainly minimizes frequency deviation, ignores frequency modulation cost, equipment load boundary and energy loss problems, and cannot support the demand for maximum regulation efficiency under complex systems. The present application constructs a multi-objective performance evaluation function system, aiming at the problem that the adjustment performance of hydroelectric generating set is easily changed by factors such as water level and water head, and integrates hydraulic constraints and energy efficiency indicators in the control strategy, considering frequency response and operation economy; the frequency steady-state index, water energy loss index and direct current capacity occupation index are fused, and an intelligent optimization algorithm suitable for nonlinear multi-constraint scene is designed, so that the operation adaptability and global effectiveness of the frequency modulation strategy are improved.

[0032] The present application breaks through the technical bottleneck of "each for itself" of multiple frequency modulation resources in AC-DC hybrid power grid, provides a systematic frequency modulation scheme with unified model, efficient control and optimization collaboration, and provides a more supportive technical foundation and regulation ability for high-proportion clean energy grid connection and power grid operation safety.

[0033] Additional aspects and advantages of the present application will become apparent in the light of the following description, or will be learned by practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0034] The above and / or additional aspects and advantages of the present application will become apparent and be readily understood from the following description, taken in conjunction with the accompanying drawings, in which: Figure 1 is a flow chart of a multi-direct current participated hydroelectric power system frequency modulation coordination optimization method of one embodiment of the present application; Figure 2 is a schematic diagram of an equivalent synchronous generator model in the multi-direct current participated hydroelectric power system frequency modulation coordination optimization method of one embodiment of the present application; Figure 3 is a schematic diagram of a unified state space model of frequency modulation resources in the multi-direct current participated hydroelectric power system frequency modulation coordination optimization method of one embodiment of the present application; Figure 4 is a schematic diagram of a parameter identification flow in the multi-direct current participated hydroelectric power system frequency modulation coordination optimization method of one embodiment of the present application; Figure 5 is a schematic diagram of a control parameter optimization flow in the multi-direct current participated hydroelectric power system frequency modulation coordination optimization method of one embodiment of the present application; Figure 6 is a schematic diagram of a rolling optimization flow in the multi-direct current participated hydroelectric power system frequency modulation coordination optimization method of one embodiment of the present application; Figure 7 is a schematic diagram of the frequency modulation relationship of frequency modulation resources in the multi-direct current participated hydroelectric power system frequency modulation coordination optimization method of one embodiment of the present application; Figure 8is a schematic diagram of active power output of step load set by a method verification in one embodiment of the present application; Figure 9 is a schematic diagram of active power output of three converter stations obtained by a method verification in one embodiment of the present application; Figure 10 is a schematic diagram of system frequency deviation by a method verification in one embodiment of the present application; Figure 11 is a schematic diagram of a module of a multi-direct-current participated hydropower system frequency modulation coordination optimization system in one embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to enable a more complete understanding of the above-mentioned objects, features and advantages of the present application, the present application will be described in further detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0036] In the following description, a large number of specific details are set forth in order to facilitate a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and therefore the scope of protection of the present application is not limited by the specific embodiments disclosed below.

[0037] The multi-direct-current participated hydropower system frequency modulation coordination optimization method and system provided by some embodiments of the present application will be described below with reference to Figures 1 to 11

[0038] Some embodiments of the present application provide a multi-direct-current participated hydropower system frequency modulation coordination optimization method.

[0039] As shown in Figure 1 , a multi-direct-current participated hydropower system frequency modulation coordination optimization method is provided in the first embodiment of the present application, which includes the following steps S1 to S4. Specifically, the disclosed method is aimed at a high proportion of hydropower, multiple HVDC channels and AGC collaborative operation scenarios, and through unified modeling, parameter identification and multi-objective optimization strategy, the system frequency control performance and frequency modulation resource allocation efficiency are improved.

[0040] S1, a unified state space model of hydropower system frequency modulation coordination optimization is established, and the unified state space model includes a hydraulic turbine governing system, a synchronous generator, a high-voltage direct-current transmission converter station frequency modulation system and an AGC control system; In some embodiments, the state space structure of the unified state space model is represented as:

[0041] ​where x represents the state variable set, including the internal state of the synchronous machine, the dynamic variable of the hydraulic turbine and the AGC integral variable, etc.; u represents the control input, including the frequency deviation, the reference voltage and the hydraulic turbine speed regulating signal, etc.; y represents the measurable output, including the frequency, the power and the opening degree, etc.; A, B, C and D constitute the system matrix, which is confirmed through subsequent parameter identification.

[0042] In one specific embodiment, the modeling of each system in the unified state space model is as follows.

[0043] According to the classical regional load frequency control model, the model of the equivalent generator and the power system is as shown in Figure 2 The specific synchronous generator adopts the classical fourth-order synchronous machine model, and the automatic voltage regulator and the power system stabilizer are considered to describe the system voltage and electromagnetic power output, which is expressed as:

[0044] wherein, represents the rotor angle of the generator, in the present disclosure, the dot mark above the parameter represents the time derivative of the parameter unless otherwise specified; represents the rated angular velocity of the rotor; represents the angular velocity of the generator; represents the inertia constant of the generator; represents the mechanical torque input, i.e. the mechanical torque received by the rotor of the generator; represents the electromagnetic torque generated by the generator; represents the damping coefficient, which is used to describe the damping effect of the rotor of the generator; represents the transient potential, i.e. the potential of the generator in the transient process; represents the time constant related to the transient potential; represents the excitation potential, i.e. the potential generated by the excitation winding of the generator; represents the time constant of the excitation system; represents the gain of the automatic voltage regulator (AVR); represents the reference voltage, i.e. the voltage value expected by the power system; represents the actual voltage.

[0045] Figure 2 wherein, represents the active power disturbance, i.e. the change of the active power in the system; represents the change of the active power of the generator, i.e. the change of the output power of the generator; s represents a complex variable in the complex frequency domain; represents the frequency change, i.e. the change of the frequency in the power system, which is used to describe the change of the guide vane opening degree.

[0046] For the hydro-turbine governing system, a third-order hydro-turbine dynamic model is adopted, considering the nonlinear coupling between water head, water flow inertia and guide vane opening control. The governor model is a PID structure with amplitude limiting and dead zone constraints, used to regulate the mechanical output power. The hydro-turbine governing system model is represented as:

[0047] wherein, represents the dynamic change rate of guide vane opening, described by the function , which depends on the change of guide vane opening and the current guide vane opening ; represents the dynamic change rate of water head, described by the function , which depends on the current guide vane opening and the current water head height ; represents the dynamic change rate of mechanical speed or power feedback, described by the function , which depends on the current water head height and the mechanical power.

[0048] By approximate modeling, the third-order model is simplified to an equivalent simplified model:

[0049] wherein, represents the transfer function of the hydro-turbine governing system; represents the adjustable regulation slope coefficient, representing the gain of the hydro-turbine; , , are time constants, used to represent the time characteristics of the hydro-turbine dynamic response.

[0050] For the frequency modulation system of the high-voltage direct current (HVDC) transmission station, the regulating behavior of the converter is constructed based on the average value modeling method. The frequency modulation control adopts a power linear regulation relationship based on the frequency deviation, represented as:

[0051] wherein, represents the power change amount of the HVDC channel represented in the Laplace transform domain; represents the adjustable regulation slope coefficient, used to adjust the response strength of the HVDC channel to the frequency deviation, supporting subsequent optimization solution; represents the time constant used to describe the dynamic characteristics of the HVDC channel regulating power change; represents the frequency deviation represented in the Laplace transform domain, i.e. the difference between the actual frequency and the rated frequency.

[0052] For the AGC control system, a PI-type automatic generation control loop is constructed using an area control error (ACE) feedback structure, which realizes the automatic distribution of active power based on the frequency and tie-line power deviation, expressed as:

[0053] wherein, represents the control output of the AGC system, usually expressed as the power set value of the generating equipment; represents the proportional coefficient of the PI control loop; represents the integral coefficient of the PI control loop; represents the area control error, representing the deviation between the actual frequency and the target frequency at time t.

[0054] Based on the above-mentioned subsystem models, a unified state space model is established as shown in Figure 3 , which integrates the synchronous generator and excitation system (integrated into the power system dynamics in the figure), AGC control system (shown as AGC system in the figure), high-voltage direct-current transmission converter station frequency modulation system (shown as HVDC system in the figure) and hydro-turbine governing system (shown as hydro-turbine-governor system in the figure) to realize the unified regulation of frequency and power in the power system. Specifically, the AGC system adjusts the output of the generator set and the HVDC system according to the frequency deviation signal to maintain the frequency stability of the system. The hydro-turbine-governor system adjusts the output power of the hydro-turbine according to the control signal of the AGC system and the load power change to meet the power demand of the system. This unified model helps to analyze and optimize the frequency regulation and power distribution strategy in the power system, and improves the stability and reliability of the system. By adjusting the parameters in the model (such as , , , etc.), the dynamic response and steady-state performance of the system can be optimized.

[0055] S2, based on real-time operation data and historical monitoring data, uses the least squares method and evolutionary algorithm fusion method to jointly identify the parameters in the unified state space model.

[0056] In some embodiments, the process of step S2 is as shown in Figure 4 , which includes: S21, obtain input data, the input data including wide area measurement system (WAMS) data during frequency modulation event, EMS (energy management system) historical data, substation SCADA (supervisory control and data acquisition system) data and PMU (synchronous phasor measurement unit) synchronous phasor data.In one specific embodiment, about 20-60 seconds of sampling data before and after the occurrence of the frequency modulation event is used as the identification window.After obtaining the input data, data cleaning and alignment are carried out, specifically, missing points can be completed by interpolation;DC component and high-frequency noise are filtered out;Savitzky-Golay filter is used to smooth the differential signal.

[0057] S22, construct an identification model, wherein the identification model extracted based on the unified state space model is represented as:

[0058] Wherein x(t) represents the system state variable (such as guide vane opening, water head, voltage, frequency, etc.) at time t; u(t) represents the control input (such as frequency deviation, AGC output, etc.) at time t; represents the parameter vector to be identified; represents the joint differential structure composed of various modeling equations.

[0059] S23, use the unit parameter manual or historical setting value as the initial reference to set the initial value of the identification parameter.

[0060] In some embodiments, a segmented time window method is used to gradually approach the dynamic segment and improve the local fitting degree.

[0061] S24, use the least square method and evolutionary algorithm joint optimization mechanism to optimize the parameters.

[0062] Specifically, the least square method is used to identify the linear time delay characteristics in the linear subsystem (such as AVR, AGC). The least square method uses OLS algorithm or RLS algorithm.

[0063] The evolutionary algorithm is used for global search for the nonlinear subsystem (such as guide vane dynamics of hydraulic turbine, frequency-power curve, etc.); the objective function of global search is:

[0064] Wherein, represents the model output; represents the actual measurement value; N represents the number of sampling points.

[0065] The evolutionary algorithm uses differential evolution algorithm (DE) or particle swarm optimization algorithm (PSO).

[0066] S25, run multiple algorithm populations (PSO, GA, etc.) in parallel in the initial stage; use local hill climbing method (such as BFGS) to refine local minimum; introduce disturbance co-evolution (such as scaling + mutation) to avoid falling into local optimum until the algorithm converges.

[0067] Specifically, the convergence criterion can be set as: The mean square error is less than a set threshold (such as The parameter variation is less than a set variation rate (such as The optimization is terminated when there is no significant improvement in continuous 5 rounds of search.

[0068] S3, based on the unified state space model and the parameter identification result, an integrated optimization objective function considering frequency deviation, water energy utilization efficiency and HVDC regulation capacity occupation is constructed, and the optimal control parameter is output.

[0069] The embodiment constructs an optimization control module for multi-objective frequency modulation performance, aiming to realize efficient collaboration among multiple frequency modulation resources, balance frequency stability, water energy utilization efficiency and HVDC channel operation economy. Based on the unified modeling and parameter identification module, the frequency deviation response, water and electricity energy consumption and HVDC regulation capacity are comprehensively considered, and the optimal frequency modulation control parameter is output through the constraint optimization method, so as to realize the performance optimization of hydropower-multi HVDC joint frequency modulation. The overall design process of the multi-objective optimization module is as shown in Figure 5

[0070] In one specific embodiment, step S3 comprises: S31, performance indicators are constructed from three dimensions of system operation performance, energy efficiency and regulation capacity, including frequency response performance indicators, water energy utilization efficiency indicators and HVDC regulation capacity occupation indicators.

[0071] Specifically, the frequency response performance indicator is used to measure the recovery ability of the system frequency deviation after disturbance response, and the integral of the regional ACE signal is taken as the measurement, which is expressed as:

[0072] Wherein, ACE represents the regional ACE signal; T represents the time period; The water energy utilization efficiency indicator is used to measure the water energy waste or non-optimal operation loss caused by frequent regulation, and an empirical model is constructed based on the water level fluctuation and power generation efficiency degradation curve for estimation; The HVDC regulation capacity occupation indicator is used to reflect the average proportion of the amplitude of the HVDC channel regulation power caused by frequency modulation, which is expressed as:

[0073] Wherein, ​represents the change amount of the HVDC channel regulation power at time t; represents the maximum regulation power amplitude 2 value of the HVDC channel.

[0074] S32, constructing a comprehensive optimization objective function based on the three performance indicators is:

[0075] wherein, represents the frequency response performance indicator; represents the water energy utilization efficiency indicator; represents the HVDC regulation capacity occupation indicator; , , All are adjustable weight parameters, which can be flexibly set according to the scene.

[0076] S33, setting the optimization variables including the water turbine speed regulator parameters (such as , , ), the HVDC regulation slope and the AGC region active frequency modulation distribution coefficient.

[0077] S34, setting the constraint conditions including the frequency constraint, the water level / water head constraint, the HVDC regulation power constraint and the dynamic response constraint.

[0078] Specifically: Frequency constraint:

[0079] Water level / water head constraint:

[0080] HVDC regulation power constraint:

[0081] Dynamic response constraint: the HVDC regulation time delay is not more than 0.2s, and the water turbine response time delay is not more than 3s.

[0082] S35, a multi-objective search method based on evolutionary strategy is adopted to construct a parallel search framework, and based on the constructed search framework, the optimal control parameters are searched under the constraint condition with the comprehensive optimization objective function as the target.

[0083] In some embodiments, the multi-objective search method adopts the NSGA-III algorithm, and the search process includes: initializing a plurality of solution vector groups; performing cross and mutation operations under the target function and the constraint; introducing a Pareto sorting and crowding distance selection mechanism; constantly evolving to generate a new solution group, and gradually approaching the Pareto optimal frontier. It should be noted that the NSGA-III algorithm is an existing algorithm, and its specific process will not be repeated here.

[0084] ​S4, based on the rolling optimization mechanism, the control parameters are issued to the hydroelectric governor, HVDC regulating device and AGC execution system to realize the dynamic collaborative scheduling of various frequency modulation resources.

[0085] In some embodiments, step S4 is based on the rolling optimization idea, and the control parameters output by the optimization module are deployed to the control terminals of the system in a periodic updating manner to realize efficient collaborative control among the hydroelectric power generation system, the HVDC power transmission system and the automatic generation control system, and to improve the closed-loop response performance of the system frequency regulation.

[0086] In one specific embodiment, the rolling optimization process is as shown in Figure 6 .

[0087] Among them, the optimization module running period can be in units of 2 minutes, and is periodically called in the energy management system of the power grid dispatching master station, combined with the latest load prediction information, unit state data and disturbance detection results, to dynamically adjust the controller parameters. After each round of optimization calculation is completed, the following key control quantities are issued to the corresponding devices: a. Turbine governor parameters: including the proportional coefficient, integral time and differential time of the PID controller, used to adjust the local response ability of the hydroelectric generating unit to the frequency deviation; b. HVDC regulation slope coefficient: determines the adjustment amplitude of the HVDC channel to the system frequency change, controls the sensitivity of its active power output to the frequency deviation; c. AGC load distribution coefficient: according to the available capacity and operating state of various resources, the load distribution proportion of hydroelectric and HVDC resources in AGC control is adjusted to realize the optimal configuration of the active control resources of the whole system.

[0088] In an AC-DC hybrid system, during primary frequency modulation, the hydroelectric power station and the HVDC converter station adjust the power output according to the optimization instructions issued by the deployment module, and then the secondary frequency modulation AGC controls the active output proportion of hydroelectric and HVDC resources. The relationship among the three is as shown in Figure 7 .

[0089] In one specific embodiment, with reference to the method of the present disclosure, a unified state space model of frequency modulation resources is built in simulation software, including hydroelectric generating units, three HVDC converter stations and AGC systems. Based on this model, model parameter identification is performed, then a multi-objective optimization function is designed, and finally a control deployment module is built.

[0090] The generator parameters, simulation working condition parameters, etc. are set, the active output of the hydroelectric generating unit and the HVDC converter station, and the system frequency deviation are simulated, and finally the effectiveness of the method is analyzed.

[0091] Figure 8The active power output of the step power consumption load is set, and the disclosure comprehensively considers factors such as frequency deviation response, water and electricity energy consumption and HVDC adjustment capacity. Considering the remaining frequency modulation capacity of the HVDC converter station, the power distribution of the HVDC converter station is effectively optimized. Figure 9 It can be seen from the active power output curve that the three HVDC converter stations will output frequency modulation power at all times. Among them, the output power of the HVDC converter station 1 is higher, the output power of the HVDC converter station 2 is lower, and the output power of the HVDC converter station 3 is the lowest. This shows that the multi-source frequency modulation integrated coordination control method in the disclosure effectively coordinates the frequency modulation capacity of the HVDC.

[0092] The multi-source frequency modulation coordination control method designed in the disclosure uses an MPC controller to adjust the system frequency. It can be seen from Figure 10 that after coordination control, the maximum output frequency deviation of the system is 0.086 Hz, and the longest recovery time after frequency fluctuation is 10.158s, which shows that the coordinated control method considering the water and electricity units, HVDC and AGC system proposed in the disclosure can quickly adjust the system frequency and ensure the stability of the system.

[0093] The application provides a kind of multi DC participation's hydroelectric system frequency modulation coordination optimization system, as shown in Figure 11 It includes: unified modeling module, parameter identification module, multi-objective optimization module and control deployment module.

[0094] The unified modeling module is used to establish a unified state space model for hydroelectric system frequency modulation coordination optimization, which includes a hydro-turbine governing system, a synchronous generator, a HVDC converter station frequency modulation system and an AGC control system. The parameter identification module uses a least squares method and an evolutionary algorithm fusion method to jointly identify the parameters in the unified state space model based on real-time operation data and historical monitoring data. The multi-objective optimization module constructs a comprehensive optimization objective function considering frequency deviation, water energy utilization efficiency and HVDC adjustment capacity occupation based on the unified state space model and parameter identification results, and outputs optimal control parameters. The control deployment module based on the rolling optimization mechanism sends control parameters to the hydroelectric governor, HVDC adjustment device and AGC execution system to realize dynamic collaborative scheduling of various frequency modulation resources.

[0095] In this specification, the illustrative descriptions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0096] Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A frequency regulation coordination optimization method for a hydropower system involving multiple DC transmission lines, characterized in that, include: A unified state-space model for frequency regulation coordination optimization of a hydropower generation system is established. The unified state-space model includes the turbine speed regulation system, synchronous generator, high-voltage direct current transmission converter station frequency modulation system, and AGC control system. Based on real-time operational data and historical monitoring data, a fusion method of least squares and evolutionary algorithm is used to jointly identify the parameters in the unified state space model. Based on the unified state-space model and parameter identification results, a comprehensive optimization objective function is constructed that considers frequency deviation, hydropower utilization efficiency and HVDC regulation capacity occupancy, and the optimal control parameters are output. Based on the rolling optimization mechanism, control parameters are sent to the hydropower speed governor, HVDC regulating device and AGC execution system to realize dynamic coordinated scheduling of various frequency regulation resources.

2. The frequency regulation coordination optimization method for a multi-DC hydropower system according to claim 1, characterized in that, The state-space structure of the unified state-space model is represented as follows: Where x represents the set of state variables, including the internal state of the synchronizing machine, the dynamic variables of the turbine, and the integral variables of AGC; u represents the control input, including the frequency deviation, the reference voltage, and the turbine speed regulation signal; y represents the measurable output, including the frequency, power, and opening degree; A, B, C, and D form the system matrix, which is identified and confirmed through parameter identification.

3. The frequency regulation coordination optimization method for a multi-DC hydropower system according to claim 2, characterized in that, The method of jointly identifying parameters in a unified state-space model using a fusion of least squares and evolutionary algorithms includes: Acquire input data, which includes wide-area measurement system data, EMS historical data, substation SCADA data, and PMU synchronization phasor data during frequency regulation events; Construct an identification model, wherein the identification model extracted based on the unified state-space model is represented as follows: Where x(t) represents the system state variable at time t; u(t) represents the control input at time t; Represents the parameter vector to be identified; This represents the joint differential structure composed of the modeling equations; Use the unit parameter manual or historical setting values ​​as initial references to identify initial parameter values; A joint optimization mechanism combining least squares method and evolutionary algorithm is employed to optimize the parameters. Specifically, least squares method is used to identify linear time-delay characteristics in linear subsystems, while evolutionary algorithm is used for global search in nonlinear subsystems. The objective function of the global search is: in, Indicates the model output; This represents the actual measured value; N represents the number of sampling points; The algorithm refines local minima using a local hill-climbing method and introduces perturbation-based co-evolution to avoid getting trapped in local optima until the algorithm converges.

4. The frequency regulation coordination optimization method for a multi-DC hydropower system according to claim 3, characterized in that, The least squares method employs either the OLS algorithm or the RLS algorithm. And / or, the evolutionary algorithm employs differential evolution or particle swarm optimization.

5. The frequency regulation coordination optimization method for a multi-DC hydropower system according to claim 3, characterized in that, When performing parameter identification, a segmented time window approach is used to approximate the dynamic segment one by one.

6. The frequency regulation coordination optimization method for a multi-DC-participating hydropower system according to claim 3, characterized in that, Based on the unified state-space model and parameter identification results, a comprehensive optimization objective function considering frequency deviation, hydropower utilization efficiency, and HVDC regulation capacity occupancy is constructed, outputting optimal control parameters, including: Performance indicators are constructed from three dimensions: system operation performance, energy efficiency, and regulation capability, including frequency response performance indicators, hydropower utilization efficiency indicators, and HVDC regulation capacity utilization indicators. The comprehensive optimization objective function is constructed based on three performance indicators as follows: in, This indicates the frequency response performance index; Indicates the efficiency of hydropower utilization; This indicates the HVDC regulation capacity utilization index; , , All are adjustable weight parameters; The optimization variables include turbine governor parameters, HVDC regulation slope, and AGC area active power frequency regulation allocation coefficient. The constraints include frequency constraints, water level / head constraints, HVDC regulation power constraints, and dynamic response constraints. A parallel search framework is constructed using a multi-objective search method based on an evolutionary strategy. Based on the constructed search framework, the optimal control parameters are searched under constraints with the goal of comprehensively optimizing the objective function.

7. The frequency regulation coordination optimization method for a hydropower system with multiple DC participants according to claim 6, characterized in that, The frequency response performance index is used to measure the system's ability to recover frequency deviation after a disturbance response, and is expressed as the integral of the regional ACE signal: in, Indicates the regional ACE signal; T represents the time period; The hydropower utilization efficiency index is used to measure the hydropower waste or suboptimal operating losses caused by frequent adjustments. An empirical model is constructed based on the water level fluctuation and power generation efficiency degradation curve for estimation. The HVDC regulation capacity occupancy index reflects the average proportion of the HVDC channel regulation power amplitude caused by frequency modulation, and is expressed as: in, This represents the change in the HVDC channel regulation power at time t. This indicates the maximum regulating power amplitude of the HVDC channel.

8. The frequency regulation coordination optimization method for a hydropower system with multiple DC participants according to claim 6, characterized in that, The multi-objective search method employs the NSGA-III algorithm, and the search process includes: Initialize a population of multiple solution vectors; Perform crossover and mutation operations under the objective function and constraints; Introduce Pareto sorting and crowd distance selection mechanisms; It continuously evolves to generate a new population of solutions, gradually approaching the Pareto optimal frontier.

9. The frequency regulation coordination optimization method for a multi-DC hydropower system according to claim 1, characterized in that, The rolling optimization mechanism, which distributes control parameters to the hydropower governor, HVDC regulator, and AGC execution system, includes: The control parameters are deployed to each control terminal of the system in real time with periodic updates. The control parameters are periodically called in the energy management system of the power grid dispatching master station, and the controller parameters are dynamically adjusted in combination with the latest load forecast information, unit status data and disturbance detection results.

10. A frequency regulation coordination optimization system for a hydropower generation system involving multiple DC transmission lines, characterized in that, include: A unified modeling module is used to establish a unified state space model for frequency regulation coordination optimization of a hydropower generation system. The unified state space model includes a turbine speed regulation system, a synchronous generator, a high-voltage direct current transmission converter station frequency modulation system, and an AGC control system. The parameter identification module uses a fusion method of least squares and evolutionary algorithm to jointly identify the parameters in the unified state space model based on real-time operation data and historical monitoring data. The multi-objective optimization module, based on a unified state-space model and parameter identification results, constructs a comprehensive optimization objective function that considers frequency deviation, hydropower utilization efficiency, and HVDC regulation capacity occupancy, and outputs the optimal control parameters. The control deployment module, based on a rolling optimization mechanism, sends control parameters to the hydropower speed governor, HVDC regulating device, and AGC execution system to achieve dynamic and coordinated scheduling of various frequency regulation resources.

Images