Self-adaptive intelligent power control method for electric energy router in photovoltaic power generation system

Abstract

The invention relates to a self-adaptive intelligent power control method for an electric energy router in a photovoltaic power generation system. According to the method, key parameters of photovoltaic, energy storage and load are monitored in real time, a virtual power pool is constructed to quantify a multi-source power margin and dynamically distribute a priority sequence, energy storage output is dynamically adjusted in combination with a feed-forward compensation control strategy, frequency modulation deviation is periodically checked, weight optimization redistribution is triggered, and energy storage output is dynamically adjusted. Cooperative control of photovoltaic load reduction frequency modulation and DC bus voltage stability is realized, the multi-source cooperative frequency modulation response capability and the energy utilization rate are effectively improved, voltage fluctuation caused by power abrupt change is suppressed, and both the system operation economy and the power supply reliability are considered.

Description

Technical Field

[0001] The present invention belongs to the technical field of power equipment, and particularly relates to an adaptive intelligent power control method for an electric energy router in a photovoltaic power generation system. Background Art

[0002] In a photovoltaic power generation system, as an energy scheduling hub, the electric energy router needs to achieve dynamic balance among photovoltaic output fluctuations, energy storage capacity limitations, and power grid frequency regulation requirements.

[0003] Traditional control methods usually control photovoltaic maximum power tracking, energy storage charge and discharge, and load management as independent modules, resulting in low multi-source cooperation efficiency. For example, when the photovoltaic system is fully generating power and the energy storage is in a high state of charge, the system cannot effectively coordinate the contradiction between light abandonment losses and frequency regulation reserve capacity, and often sacrifices the utilization rate of some renewable energy by forced load reduction or direct light abandonment; at the same time, there is a lack of effective decoupling means for the coupling problem between frequency regulation response and DC bus voltage stability, and power mutations are likely to cause voltage instability.

[0004] In addition, existing solutions do not make full use of adjustable resources on the load side, and the economic weight distribution is fixed, making it difficult to adapt to the dynamic changes of real-time electricity prices and equipment health status.

[0005] The above problems restrict the frequency regulation ability and operation economy of the photovoltaic and energy storage system in high-penetration scenarios. There is an urgent need for an adaptive control mechanism that can penetrate multi-source power margin quantification, dynamic priority decision-making, and rapid power compensation. Summary of the Invention

[0006] Based on this, in view of the above technical problems, it is necessary to provide an adaptive intelligent power control method for an electric energy router in a photovoltaic power generation system.

[0007] The present application provides an adaptive intelligent power control method for an electric energy router in a photovoltaic power generation system, including: S1: Calculate the photovoltaic load reduction margin, energy storage discharge margin, and load curtailment margin according to the output power of the photovoltaic array, the state of charge of the energy storage system, and the power of the adjustable load; S2: Calculate the total virtual reserve according to the photovoltaic load reduction margin, energy storage discharge margin, and load curtailment margin; S3: According to the total virtual reserve, combined with the type of frequency regulation demand and dynamic economic weight, allocate the priority sequences of the photovoltaic, energy storage, and load; S4: Based on the priority sequence, adjust the output of the energy storage system through a feed-forward compensation control strategy to suppress the disturbance of the photovoltaic load reduction power to the DC bus voltage; S5: Execute multi-source collaborative frequency modulation according to the priority sequence and the total virtual reserve; periodically check the frequency modulation power deviation based on the real-time volatility of the DC bus voltage and the rate of change of the state of charge of the energy storage system, and adjust the dynamic economic weight according to the frequency modulation power deviation.

[0008] In one embodiment, step S1 includes: S11: Track the maximum power point of the photovoltaic array based on the incremental conductance method to obtain the maximum available power of the photovoltaic, and calculate the photovoltaic load reduction margin according to the formula ; where is the photovoltaic load reduction margin, is the maximum available power of the photovoltaic, is the output power of the photovoltaic array; S12: Calculate the charge and discharge power limit of the energy storage system according to the health state of the energy storage system according to the following formula: ; where α and β are configuration parameters preset according to the energy storage type of the energy storage system, is the charge and discharge power limit of the energy storage system, is the rated charge and discharge power of the energy storage system, is the health state of the energy storage system, is the reference value of the energy storage health state; S13: Calculate the energy storage discharge margin based on the charge and discharge power limit and the state of charge of the energy storage system according to the following formula: ; where is the energy storage discharge margin, is the state of charge of the energy storage system, is the highest state of charge allowed by the energy storage system; S14: Identify the non-critical loads in the adjustable load through the load priority classification model to obtain the power of the non-critical loads, and calculate the load reduction margin according to the following formula: ; where is the load reduction margin, is the power of the non-critical load, and γ is a preset load reduction coefficient.

[0009] In one embodiment, calculating the total virtual reserve includes: Perform weighted fusion on the photovoltaic load reduction margin, the energy storage discharge margin, and the load reduction margin to obtain the total virtual reserve; the calculation formula for weighted fusion is: ; where is the total virtual reserve, is the energy storage discharge efficiency coefficient of the energy storage system, is the load shedding efficiency coefficient.

[0010] In one embodiment, before calculating the total virtual reserve, it further includes: S21: Monitor the temperature of the energy storage system. If the temperature exceeds the preset temperature threshold, reduce the energy storage discharge efficiency coefficient according to a linear relationship; S22: Modify the energy storage discharge efficiency coefficient according to the number of cycles of the energy storage system according to an exponential decay model; S23: Substitute the modified energy storage discharge efficiency coefficient into the calculation of the total virtual reserve.

[0011] In one embodiment, step S3 includes: S31: Set the frequency modulation emergency weight according to the type of frequency modulation demand; S32: Construct an economic cost function based on the real-time electricity price and loss cost; the expression of the economic cost function is: ; where, is the dynamic economic weight calculated according to the real-time electricity price and loss cost; S33: Input the total virtual reserve, the frequency modulation emergency weight, and the economic cost function into a preset fuzzy logic controller to generate a priority sequence through a preset rule base.

[0012] In one embodiment, step S4 includes: S41: Determine whether to preferentially call the photovoltaic load shedding margin according to the priority sequence; if called, calculate the photovoltaic power adjustment amount according to the frequency modulation demand and the photovoltaic load shedding margin, and generate the energy storage feedforward reference power through the photovoltaic power adjustment amount and the feedforward compensation coefficient; the calculation formula for generating the energy storage feedforward reference power is: ; where, is the energy storage feedforward reference power; is the photovoltaic power adjustment amount and satisfies ; is the feedforward compensation coefficient, , is the DC bus capacitance value, is the control delay; S42: Adjust the deviation of the DC bus voltage through a voltage outer loop PI controller to generate the energy storage feedback reference power; the calculation formula for the energy storage feedback reference power is: ; where, is the reference power of energy storage feedback, is the DC bus voltage deviation at moment, represents the cumulative DC bus voltage deviation over time, and is the proportional-integral parameter set according to the dynamic response requirements of the energy storage system; S43: Superimpose the energy storage feedforward reference power and the energy storage feedback reference power to obtain the total energy storage reference power, and control the energy storage system to output power according to the total energy storage reference power.

[0013] In one embodiment, step S5 includes: S51: Allocate the PV curtailment margin, the energy storage discharge margin, and the load shedding margin to the grid frequency regulation demand according to the priority sequence to generate the actual frequency regulation power; S52: Calculate the target deviation between the actual frequency regulation power and the target frequency regulation demand. If the ratio of the target deviation to the target frequency regulation demand is greater than the preset ratio, reallocate the dynamic economic weight based on the preset rules; S53: Based on the real-time volatility of the DC bus voltage and the state of charge change rate of the energy storage system, correct the dynamic economic weight through the gradient descent algorithm.

[0014] The above-mentioned adaptive intelligent power control method for the power router in a photovoltaic power generation system monitors the key parameters of PV, energy storage, and load in real time, constructs a virtual power pool to quantify the multi-source power margin and dynamically allocate the priority sequence, combines the feedforward compensation control strategy to dynamically adjust the energy storage output, and at the same time periodically checks the frequency regulation deviation and triggers the weight optimization and reallocation, realizing the coordinated control of PV curtailment frequency regulation and DC bus voltage stability, effectively improving the multi-source coordinated frequency regulation response ability and energy utilization rate, suppressing the voltage fluctuation caused by power mutation, and taking into account the system operation economy and power supply reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the following will briefly introduce the drawings required for the description of the embodiments or the related art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 is a schematic flowchart of an adaptive intelligent power control method for a power router in a photovoltaic power generation system provided by the present invention; Figure 2 is a schematic flowchart of dynamically correcting the energy storage discharge efficiency coefficient in an alternative embodiment of the present invention; Figure 3 Schematic diagram of the process for generating a priority sequence in an alternative embodiment of the present invention; Figure 4 Schematic diagram of the process for controlling the output of an energy storage system in an alternative embodiment of the present invention; Figure 5 Schematic diagram of the process for performing coordinated frequency regulation and adjusting the dynamic economic weight in an alternative embodiment of the present invention. Detailed implementation manners

[0017] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0018] Refer to Figure 1 , which shows a schematic diagram of the process of an adaptive intelligent power control method for an electric energy router in a photovoltaic power generation system provided by the present application. The method includes the following steps: S1: Calculate the photovoltaic derating margin, the energy storage discharge margin, and the load curtailment margin according to the output power of the photovoltaic array, the state of charge of the energy storage system, and the power of the adjustable load.

[0019] Specifically, in a photovoltaic power generation system, the electric energy router, as the core energy scheduling hub, its primary task is to evaluate the available power margins of various resources in the system in real time. The power margin is the power capacity that the system can provide additionally while meeting the current operating requirements, and it is the basis for realizing dynamic power distribution and frequency regulation control.

[0020] The photovoltaic derating margin refers to the redundant power that the photovoltaic array can release under the current illumination conditions. The calculation of this margin is based on the difference between the theoretical maximum power of the maximum power point tracking (MPPT) of the photovoltaic array and the actual output power. The specific implementation means can be: real-time tracking of the maximum power point of the photovoltaic array through the MPPT algorithm to ensure that the photovoltaic array always operates at the maximum power output state under different illumination and temperature conditions. At the same time, the system monitors the actual output power of the photovoltaic array in real time through a power sensor. By comparing the maximum power tracked by the MPPT with the actual output power, the redundant power of the photovoltaic array is calculated. This redundant power reflects the additional power that the photovoltaic system can release without reducing the power generation efficiency. The calculation of the redundant power takes into account the illumination intensity, temperature changes, and the aging characteristics of the photovoltaic array. By real-time monitoring of the environmental parameters, the evaluation result of the redundant power is dynamically adjusted.

[0021] The energy storage dischargeable margin refers to the maximum dischargeable power that an energy storage system can provide at the current state of charge (SOC). The evaluation of this margin comprehensively considers the state of charge, charge-discharge efficiency, and health state of the energy storage system. The specific implementation means can be as follows: The battery management system (BMS) is used to monitor the state of charge (SOC) of the energy storage system in real time to ensure that the SOC value is within a safe range. The charge-discharge efficiency of the energy storage system is affected by battery type, temperature, and charge-discharge rate. Through experimental data and model prediction, the charge-discharge efficiency of the energy storage system is dynamically evaluated. The health state (SOH) of the energy storage system directly affects its available power. By monitoring parameters such as the internal resistance change and capacity attenuation of the battery, the health state of the energy storage system is evaluated, and the calculation of the dischargeable margin is adjusted accordingly. Combining SOC, charge-discharge efficiency, and SOH, the system dynamically evaluates the maximum dischargeable power of the energy storage system under the current state.

[0022] The load curtailment margin refers to the power that the system can release by curtailing flexible loads under the frequency regulation demand. The evaluation of this margin considers the type, response speed, and elasticity coefficient of the load. The specific implementation means can be as follows: The loads are classified into rigid loads (such as lighting and air conditioning) and flexible loads (such as electric vehicle charging and energy storage devices). Each load type has a corresponding elasticity coefficient, which reflects the proportion of power that can be curtailed. The flexible loads are grouped by priority through a hierarchical clustering algorithm. The loads with higher priority are curtailed first under the frequency regulation demand to minimize the impact on users. Through historical data and real-time monitoring, the response ability of the flexible loads is evaluated. The loads with stronger response ability can curtail power faster under the frequency regulation demand. Combining the elasticity coefficient, priority, and response ability, the load curtailment margin is dynamically calculated.

[0023] S2: Calculate the total virtual reserve according to the photovoltaic power reduction margin, energy storage dischargeable margin, and load curtailment margin.

[0024] Specifically, the total virtual reserve is constructed by dynamically weighted fusion of the power margins of various resources, and is used to characterize the comprehensive response ability of the system under the frequency regulation demand.

[0025] The total virtual reserve can be constructed by weighted fusion of the photovoltaic power reduction margin, energy storage dischargeable margin, and load curtailment margin. The weight coefficient reflects the relative importance of each resource in frequency regulation.

[0026] The weight coefficient can be dynamically adjusted according to factors such as real-time electricity price fluctuations, energy storage health state, and load response delay. For example, when the real-time electricity price is high, the weight of the energy storage system will be reduced to reduce the use of the energy storage system and lower the operating cost. By weighted fusion of the power margins of various resources, the total virtual reserve is constructed. This total not only reflects the physical power margin of the system, but also comprehensively considers factors such as economy and response speed.

[0027] The real-time electricity price fluctuation has a direct impact on the economy of the system. The economic weight can be dynamically adjusted by the following means: monitor the electricity price fluctuation of the power grid in real time, and predict the electricity price trend in the future for a period of time through the electricity price prediction algorithm. Combine the electricity price fluctuation and the cycle life attenuation cost of the energy storage system to dynamically adjust the economic weights of various resources. For example, when the electricity price is high, load curtailment is preferred over energy storage discharge to reduce the operating cost.

[0028] The health state of the energy storage system directly affects its available power and economy. The health state of the energy storage is evaluated by the following means: monitor parameters such as the internal resistance change and capacity attenuation of the battery to evaluate the health state of the energy storage system. Calculate its life attenuation cost according to the health state and charge-discharge times of the energy storage system. This cost is used to adjust the economic weight of the energy storage system to ensure that resources with less impact on the life of the energy storage system are preferred under the frequency regulation demand.

[0029] S3: According to the total virtual reserve, combine the type of frequency regulation demand and the dynamic economic weight to allocate the priority sequences of photovoltaic, energy storage and load.

[0030] Specifically, under the frequency regulation demand, the priority sequence of resources is dynamically adjusted according to the total virtual reserve and the dynamic economic weight.

[0031] The frequency regulation demand can be divided into the following three categories according to the grid frequency deviation and its change rate: 1) Second-level rapid response: When the grid frequency deviation is large and the change rate is high, the system needs to respond quickly within seconds, and energy storage systems and load curtailment are preferred.

[0032] 2) Minute-level deep compensation: When the grid frequency deviation is large but the change rate is low, the system needs to perform deep compensation within minutes, and photovoltaic power reduction and energy storage systems are preferred.

[0033] 3) Hour-level economic dispatch: When the grid frequency deviation is small and the change rate is low, the system needs to perform economic dispatch within hours, and load curtailment and photovoltaic power reduction are preferred.

[0034] The economic weight can be dynamically adjusted according to the real-time electricity price, the cycle life attenuation cost of the energy storage and the curtailment penalty coefficient. The specific implementation means can be as follows: 1) Impact of real-time electricity price: The real-time electricity price fluctuation directly affects the economic weights of various resources. When the electricity price is high, load curtailment is preferred over energy storage discharge to reduce the operating cost.

[0035] 2) Energy storage life attenuation cost: Calculate its life attenuation cost according to the health state and charge-discharge times of the energy storage system. This cost is used to adjust the economic weight of the energy storage system to ensure that resources with less impact on the life of the energy storage system are preferred under the frequency regulation demand.

[0036] 3) Curtailment penalty coefficient: The curtailment penalty coefficient reflects the economic impact of curtailment losses. By dynamically adjusting the curtailment penalty coefficient, the priority of PV load reduction is optimized.

[0037] The priority sequence is generated by an improved path optimization algorithm, and the specific implementation means can be as follows: 1) Resource topology map construction: The system constructs a resource topology map, where the nodes in the map represent different resources (PV, energy storage, load), and the edges represent the power distribution relationship between resources.

[0038] 2) Path optimization algorithm: An improved Dijkstra algorithm is used to find the optimal power distribution path in the resource topology map. The algorithm comprehensively considers the frequency regulation requirements, economic weight, and response speed to generate a resource priority sequence that meets the frequency regulation requirements.

[0039] 3) Dynamic adjustment: The priority sequence is dynamically adjusted according to the real-time working conditions to ensure that the system always operates with the optimal resource allocation scheme under different frequency regulation requirements.

[0040] S4: Based on the priority sequence, the output of the energy storage system is adjusted through a feedforward compensation control strategy to suppress the disturbance of the PV load reduction power on the DC bus voltage.

[0041] Specifically, for the DC bus voltage disturbance caused by PV power fluctuations, the system uses a feedforward compensation control strategy to suppress it.

[0042] PV power fluctuations will cause DC bus voltage disturbances. The disturbances can be predicted by the following means: The change rate of PV output power is monitored in real time through a power sensor to predict the trend of power fluctuations. The change rate of power is filtered by a Kalman filter to improve the accuracy of prediction. Considering the response delay of the energy storage system, the actual amplitude and time of voltage disturbances are predicted through a delay compensation algorithm.

[0043] The output power of the energy storage system is adjusted through feedforward compensation control. The specific implementation means can be: A feedforward compensation term is superimposed on the output power of the energy storage system. By dynamically adjusting the output of the energy storage system, the disturbance of PV power fluctuations on the bus voltage is suppressed. The sensitivity of the compensation control is adjusted through a dynamic damping coefficient to ensure that the bus voltage fluctuation is controlled within a reasonable range. The compensation control ensures that the DC bus voltage fluctuation is controlled within an appropriate voltage range, improving the stability and reliability of the system.

[0044] S5: According to the priority sequence and the total virtual reserve, multi-source collaborative frequency regulation is performed; based on the real-time volatility of the DC bus voltage and the change rate of the state of charge of the energy storage system, the frequency regulation power deviation is periodically checked, and the dynamic economic weight is adjusted according to the frequency regulation power deviation.

[0045] Specifically, the system performs multi-source collaborative frequency regulation based on the priority sequence and the total virtual reserve, and dynamically adjusts the economic weight by periodically checking the frequency regulation power deviation.

[0046] The system distributes the frequency regulation power according to the priority sequence. The specific implementation means can be: sequentially distributing the frequency regulation power according to the priority sequence. Resources with higher priorities participate in frequency regulation first to ensure a rapid response to frequency regulation demands. Each resource adjusts its power according to the allocated frequency regulation power. The energy storage system adjusts its power through charge and discharge, the photovoltaic system adjusts its power through load reduction, and the load adjusts its power through curtailment.

[0047] By real-time monitoring the volatility of the DC bus voltage and the state of charge change rate of the energy storage system, the frequency regulation power deviation is periodically checked. The specific implementation means can be: real-time monitoring the volatility of the DC bus voltage through a voltage sensor to evaluate the effect of frequency regulation control. Real-time monitoring the state of charge change rate of the energy storage system through the BMS to evaluate the frequency regulation contribution of the energy storage system. By comparing the actual frequency regulation power with the target frequency regulation power, the frequency regulation power deviation is calculated. Deviation checking ensures the accuracy and reliability of frequency regulation control. According to the checking results, the economic weight is dynamically adjusted to optimize the resource utilization efficiency. For example, when the state of charge change rate of the energy storage system is too high, its economic weight is reduced, its frequency regulation contribution is decreased, and its service life is extended.

[0048] The above-mentioned adaptive intelligent power control method for the power router in a photovoltaic power generation system, by real-time monitoring the key parameters of the photovoltaic, energy storage, and load, constructs a virtual power pool to quantify the multi-source power margin and dynamically allocate the priority sequence, combines the feed-forward compensation control strategy to dynamically adjust the energy storage output, and at the same time periodically checks the frequency regulation deviation and triggers the weight optimization and reallocation, realizes the coordinated control of photovoltaic load reduction frequency regulation and DC bus voltage stability, effectively improves the multi-source collaborative frequency regulation response ability and energy utilization rate, suppresses the voltage fluctuation caused by power mutation, and takes into account the system operation economy and power supply reliability.

[0049] In an alternative embodiment, S1 includes the following steps: S11: Track the maximum power point of the photovoltaic array based on the incremental conductance method to obtain the maximum available power of the photovoltaic, and calculate the photovoltaic load reduction margin according to the formula ; where, is the photovoltaic load reduction margin, is the maximum available power of the photovoltaic, is the output power of the photovoltaic array.

[0050] Specifically, the incremental conductance method is a commonly used MPPT algorithm that locates the maximum power point by monitoring the conductance change of the photovoltaic array. When there is a difference between the output power of the photovoltaic array and the theoretical power at the maximum power point, this difference is the photovoltaic load shedding margin. Specifically, by calculating the maximum power that can be generated by the photovoltaic array under the current light conditions and the actual output power , the photovoltaic load shedding margin is obtained . This margin reflects the additional power that can be released without affecting the normal operation of the photovoltaic system

[0051] S12: According to the state of health of the energy storage system, calculate the charge and discharge power limit of the energy storage system according to the following formula ; where α and β are configuration parameters preset according to the energy storage type of the energy storage system is the charge and discharge power limit of the energy storage system is the rated charge and discharge power of the energy storage system is the state of health of the energy storage system is the reference value of the energy storage health state

[0052] Specifically, the state of health (SOH) of the energy storage system directly affects its charge and discharge capacity. To ensure the energy storage system operates in an efficient and safe state, the charge and discharge power limit is dynamically adjusted according to its state of health. Through the formula , combined with the rated power of the energy storage system and the state of health (SOH), the current charge and discharge power limit is calculated. The configuration parameters α and β in this formula are preset according to the energy storage type, and they ensure that different types of energy storage systems can operate within their optimal working ranges. In this way, the system can maximize its role in frequency modulation and power support while protecting the energy storage device

[0053] S13: Based on the charge and discharge power limit and the state of charge of the energy storage system, calculate the energy storage discharge margin according to the following formula ; where is the energy storage discharge margin is the state of charge of the energy storage system is the highest state of charge allowed for the energy storage system

[0054] Specifically, after determining the charge and discharge power limit of the energy storage system, combined with the current state of charge (SOC) of the energy storage system, the discharge margin This margin represents the maximum discharge power that the energy storage system can provide under the current state. During the calculation process, the highest allowable state of charge of the energy storage system is considered , to ensure that the energy storage system will not be over-discharged, thereby protecting the battery life and performance. In this way, the system can understand the available power of the energy storage system in real time.

[0055] S14: Identify the non-critical loads in the adjustable loads through the load priority classification model, obtain the power of the non-critical loads, and calculate the load curtailment margin according to the following formula: ; where is the load curtailment margin, is the power of the non-critical loads, and γ is the preset load reduction coefficient.

[0056] Specifically, to manage the load more effectively, the system identifies the non-critical loads in the adjustable loads through the load priority classification model. The non-critical loads can be curtailed under the grid frequency regulation demand to release power. By calculating the power of these non-critical loads and the product sum of the preset load reduction coefficient , the load curtailment margin is obtained. This margin reflects the power that the system can release by curtailing non-critical loads without affecting critical loads, providing additional flexibility for frequency regulation and power balance.

[0057] In an optional embodiment, the specific steps for calculating the total virtual reserve are as follows: Perform weighted fusion on the PV curtailment margin, the energy storage discharge margin, and the load curtailment margin to obtain the total virtual reserve; the calculation formula for weighted fusion is: ; where is the total virtual reserve, is the energy storage discharge efficiency coefficient of the energy storage system, is the load reduction efficiency coefficient.

[0058] Specifically, the total virtual reserve is obtained by performing weighted fusion on the PV curtailment margin, the energy storage discharge margin, and the load curtailment margin. This total not only reflects the physical power margin of the system but also comprehensively considers the efficiency and economy of each resource in frequency regulation.

[0059] The calculation of the total virtual reserve adopts the method of weighted fusion, which can comprehensively consider the contribution degree and efficiency of different resources in frequency regulation. By assigning different weights to the power margins of each resource, the system can more accurately evaluate the overall frequency regulation ability.

[0060] The photovoltaic available load reduction margin directly reflects the redundant power that the photovoltaic system can release under current conditions. When calculating the total virtual reserve, this margin is directly included because the frequency regulation response speed of the photovoltaic system is fast and there is no additional energy loss.

[0061] The discharge efficiency of the energy storage system has a direct impact on the actual available power. Therefore, when calculating the total virtual reserve, the available discharge margin of the energy storage is multiplied by the discharge efficiency coefficient of the energy storage system to ensure that the total virtual reserve can truly reflect the actual contribution of the energy storage system in frequency regulation.

[0062] Although load shedding can quickly respond to the frequency regulation demand, it will also cause certain efficiency losses. Therefore, the available load shedding margin is multiplied by the load reduction efficiency coefficient when calculating the total virtual reserve to reflect the actual effect of load shedding in frequency regulation.

[0063] Reference Figure 2 In an optional embodiment, before calculating the total virtual reserve, the following steps are further included: S21: Monitor the temperature of the energy storage system. If the temperature exceeds the preset temperature threshold, reduce the energy storage discharge efficiency coefficient according to a linear relationship.

[0064] Specifically, the temperature of the energy storage system has a significant impact on its performance and efficiency. High temperature will accelerate the internal chemical reaction of the battery, resulting in capacity attenuation and increased internal resistance, thereby reducing the discharge efficiency. Therefore, real-time monitoring of the temperature of the energy storage system is the key to ensuring its efficient operation. The system can preset a temperature threshold. When the temperature of the energy storage system exceeds this threshold, it indicates that the system may be in an overheated state and measures need to be taken to reduce the efficiency coefficient to protect the battery. The linear relationship between temperature and the efficiency coefficient can be determined through experimental data and model analysis. When the temperature exceeds the threshold, the system reduces the energy storage discharge efficiency coefficient according to this linear relationship to ensure that under high temperature conditions, the calculation of the total virtual reserve can truly reflect the actual available power of the energy storage system.

[0065] S22: Modify the energy storage discharge efficiency coefficient according to the number of cycles of the energy storage system according to an exponential decay model.

[0066] Specifically, the number of cycles (charge-discharge cycles) of the energy storage system directly affects its health state and discharge efficiency. As the number of cycles increases, the battery capacity gradually decays, and the internal resistance gradually increases, resulting in a decrease in discharge efficiency. An exponential decay model between the number of cycles and the efficiency coefficient can be established through experiments and data analysis. This model can accurately describe the decreasing trend of the efficiency coefficient as the number of cycles increases. The system real-time monitors the number of cycles of the energy storage system and dynamically corrects the energy storage discharge efficiency coefficient according to the exponential decay model. This correction ensures that the calculation of the total virtual reserve can consider the aging effect of the battery and improve the accuracy of the calculation results.

[0067] S23: Substitute the corrected energy storage discharge efficiency coefficient into the calculation of the total virtual reserve.

[0068] Specifically, after completing the temperature monitoring and the number of cycles monitoring, the system updates the corrected energy storage discharge efficiency coefficient to the calculation model of the total virtual reserve. By introducing the corrected efficiency coefficient, the calculation of the total virtual reserve can more accurately reflect the actual available power of the energy storage system in the current state. This helps to improve the frequency regulation ability and operation economy of the system. The corrected efficiency coefficient not only improves the calculation accuracy of the total virtual reserve, but also provides more reliable data support for subsequent power distribution and frequency regulation control. In this way, the system can dynamically optimize resource allocation, extend the service life of the energy storage system, and improve the performance and reliability of the entire photovoltaic power generation system.

[0069] Reference Figure 3 , in an optional embodiment, S3 includes the following steps: S31: Set the frequency regulation emergency weight according to the type of frequency regulation demand.

[0070] Specifically, the frequency regulation demand can be divided into three categories: second-level rapid response, minute-level deep compensation, and hour-level economic dispatch according to the grid frequency deviation and its change rate. Each type corresponds to a different level of urgency.

[0071] According to the urgency of the frequency regulation demand, the system dynamically sets the frequency regulation emergency weight. For example, the emergency weight of the second-level rapid response demand is the highest, the minute-level deep compensation is the second, and the hour-level economic dispatch is the lowest. This weight setting ensures that in an emergency, the system can respond quickly and preferentially use resources with fast response speeds (such as the energy storage system).

[0072] S32: Construct an economic cost function based on the real-time electricity price and the loss cost; the expression of the economic cost function is: ; where is the dynamic economic weight calculated according to the real-time electricity price and the loss cost.

[0073] Specifically, the real-time electricity price reflects the price fluctuations in the current electricity market, and the loss cost includes the life attenuation cost of the energy storage system, the curtailment loss cost of the photovoltaic system, and the cost of reduced user satisfaction caused by load shedding.

[0074] The system dynamically calculates the economic weights based on the real-time electricity price and the loss cost . These weights reflect the usage costs of different resources under the current economic conditions. For example, when the real-time electricity price is high, the economic weight of the energy storage system will be reduced to decrease its usage and lower the operating cost.

[0075] By linearly combining the power margins of the photovoltaic, energy storage, and load and their corresponding economic weights, an economic cost function is constructed. This function is used to evaluate the economy of different resources in frequency regulation, ensuring that the system optimizes the operating cost while meeting the frequency regulation requirements.

[0076] S33: Input the total virtual reserve, the frequency regulation emergency weight, and the economic cost function into a preset fuzzy logic controller, and generate a priority sequence through a preset rule base.

[0077] Specifically, the total virtual reserve, the frequency regulation emergency weight, and the economic cost function are used as input parameters and input into a preset fuzzy logic controller. The fuzzy logic controller processes the input parameters through a fuzzy inference system and generates a priority sequence according to the preset rule base. The rule base contains a series of fuzzy rules, such as "if the frequency regulation demand is urgent and the economic cost is high, then the energy storage system is preferentially used". Through fuzzy inference and defuzzification processing, the fuzzy logic controller outputs a specific priority sequence. This sequence determines the usage order of each resource in frequency regulation, ensuring that the system optimizes the economy and response speed while meeting the frequency regulation requirements.

[0078] Reference Figure 4 , in an optional embodiment, S4 includes the following steps: S41: Determine whether to preferentially call the photovoltaic load reduction margin according to the priority sequence; if so, calculate the photovoltaic power adjustment amount based on the frequency regulation demand and the photovoltaic load reduction margin, and generate the energy storage feedforward reference power through the photovoltaic power adjustment amount and the feedforward compensation coefficient; the calculation formula for generating the energy storage feedforward reference power is: ; Wherein, is the energy storage feedforward reference power; is the photovoltaic power adjustment amount, and satisfies ; is the feedforward compensation coefficient, , is the DC bus capacitance value, is the control delay.

[0079] Specifically, according to the priority sequence, the system first determines whether to preferentially call the photovoltaic load reduction margin. If called, the system will enter the photovoltaic power adjustment process.

[0080] Calculate the photovoltaic power adjustment amount based on the frequency regulation demand and the photovoltaic load reduction margin . This adjustment amount reflects the power that the photovoltaic system needs to release to meet the frequency regulation demand.

[0081] Feedforward compensation coefficient According to the DC bus capacitance value and the control delay Calculate. This coefficient is used to compensate for the impact of control delay and ensure the accuracy of the feedforward reference power.

[0082] Generate the energy storage feedforward reference power by multiplying the photovoltaic power adjustment amount by the feedforward compensation coefficient . This reference power is used to guide the power output of the energy storage system in the feedforward control link to ensure a fast response to photovoltaic power fluctuations.

[0083] S42: Adjust the deviation of the DC bus voltage through the voltage outer loop PI controller to generate the energy storage feedback reference power; the calculation formula for the energy storage feedback reference power is: ; Where, is the energy storage feedback reference power, is the DC bus voltage deviation at moment, represents the accumulation of the DC bus voltage deviation over time, and are the proportional-integral parameters set according to the dynamic response requirements of the energy storage system.

[0084] Specifically, the system monitors the deviation of the DC bus voltage in real time through a voltage sensor , and this deviation reflects the difference between the current bus voltage and the target voltage.

[0085] The parameters of the proportional-integral (PI) controller are set according to the dynamic response requirements of the energy storage system. These parameters determine the adjustment intensity and speed of the controller for voltage deviation.

[0086] The integral term functions to adjust the energy storage feedback reference power according to the accumulation of voltage deviation over a past period of time to reduce the voltage deviation and improve the stability of the system. If the voltage deviation persists, the integral term will gradually accumulate, prompting the energy storage system to output more power to correct the voltage deviation.

[0087] The voltage deviation is adjusted proportionally and integrally by a PI controller to generate the energy storage feedback reference power. This reference power is used to guide the power output of the energy storage system in the feedback control link to ensure that the bus voltage is stabilized within the target range.

[0088] S43: Superimpose the energy storage feedforward reference power and the energy storage feedback reference power to obtain the total energy storage reference power, and control the energy storage system to output power according to the total energy storage reference power.

[0089] Specifically, superimpose the energy storage feedforward reference power and the energy storage feedback reference power to obtain the total energy storage reference power, ensuring the combination of the fast response of the feedforward control and the precise regulation of the feedback control.

[0090] According to the total energy storage reference power, control the energy storage system to output the corresponding power, ensuring that while the energy storage system meets the frequency modulation requirements, it suppresses the disturbance of the photovoltaic power fluctuation to the DC bus voltage, and improves the stability and reliability of the system.

[0091] Reference Figure 5 , in an optional embodiment, S5 includes the following steps: S51: Allocate the photovoltaic load reduction margin, the energy storage discharge margin, and the load curtailment margin to the grid frequency modulation demand according to the priority sequence to generate the actual frequency modulation power.

[0092] Specifically, according to the priority sequence, the system sequentially calls the photovoltaic load reduction margin, the energy storage discharge margin, and the load curtailment margin to meet the grid frequency modulation demand. Resources with higher priority participate in frequency modulation first to ensure a fast response to the frequency modulation demand.

[0093] The system dynamically allocates the actual frequency modulation power according to the magnitude of the frequency modulation demand and the available margins of each resource. For example, if the frequency modulation demand is small, it may only be necessary to call the photovoltaic load reduction margin; if the frequency modulation demand is large, it is necessary to sequentially call the energy storage discharge margin and the load curtailment margin.

[0094] Through the above allocation mechanism, the system generates the actual frequency modulation power. This power reflects the frequency modulation support that the system can provide under the current working conditions to ensure that the grid frequency is stabilized within the target range.

[0095] S52: Calculate the target deviation between the actual frequency modulation power and the target frequency modulation demand. If the ratio of the target deviation to the target frequency modulation demand is greater than the preset ratio, reallocate the dynamic economic weight based on the preset rules.

[0096] Specifically, the system calculates the deviation between the actual frequency regulation power and the target frequency regulation demand in real time. This deviation reflects the accuracy of the system in meeting the frequency regulation demand. The ratio of the target deviation to the target frequency regulation demand is compared with a preset ratio. If this ratio exceeds the preset value, it indicates that there is a large error in the frequency regulation process of the system, and it is necessary to redistribute the economic weights. According to the preset rules, the system redistributes the dynamic economic weights. For example, if the deviation is large, it may be necessary to increase the weight of the energy storage system to improve its frequency regulation contribution and reduce the deviation.

[0097] S53: Based on the real-time volatility of the DC bus voltage and the state-of-charge change rate of the energy storage system, the dynamic economic weights are corrected by the gradient descent algorithm.

[0098] Specifically, the system monitors the volatility of the DC bus voltage and the state-of-charge change rate of the energy storage system in real time. These parameters reflect the current operating state of the system and the usage of the energy storage system. Through the gradient descent algorithm, the system corrects the dynamic economic weights according to the real-time parameters. The gradient descent algorithm optimizes through iteration and gradually adjusts the weights to minimize the target deviation. The corrected economic weights can better reflect the frequency regulation effect and economy of each resource under the current working conditions, ensuring that the system optimizes the operating cost while meeting the frequency regulation demand.

[0099] The above-mentioned adaptive intelligent power control method for the power router in a photovoltaic power generation system constructs a virtual power pool to integrate the multi-source power margin by real-time monitoring of the dynamic parameters of the photovoltaic, energy storage, and load, generates a priority sequence based on fuzzy logic and dynamic economic weights, combines the feed-forward compensation coefficient to calibrate and control the dynamic regulation of the energy storage output to offset the photovoltaic power mutation. At the same time, the weight allocation is periodically optimized through the gradient descent algorithm and the load adjustable resources are linked, forming a full-process adaptive mechanism of "power margin quantification - dynamic decision-making - collaborative control - closed-loop feedback", finally realizing the efficient coordination of the frequency regulation demand and the bus voltage stability, significantly improving the multi-source collaborative frequency regulation response speed and energy utilization rate, avoiding the risks of light abandonment and energy storage overload, and at the same time optimizing the system economy and power supply reliability.

[0100] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown sequentially according to the indications of the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless there is a clear indication in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0101] The above-described embodiments only express several implementation manners of the embodiments of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent for the embodiments of the application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the embodiments of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the embodiments of the present application.

Claims

1. An adaptive intelligent power control method for an electrical energy router in a photovoltaic power generation system, characterized in that, The method includes: S1: Calculate the photovoltaic (PV) load reduction margin, energy storage discharge margin, and load curtailment margin according to the output power of the PV array, the state of charge (SOC) of the energy storage system, and the power of the adjustable load. S2: Calculate the total virtual reserve according to the PV load reduction margin, the energy storage discharge margin, and the load curtailment margin. S3: Allocate the priority sequences of PV, energy storage, and load according to the total virtual reserve, in combination with the frequency regulation demand type and the dynamic economic weight. S4: Based on the priority sequences, adjust the output of the energy storage system through a feed-forward compensation control strategy to suppress the disturbance of the PV load reduction power on the DC bus voltage. S5: Perform multi-source collaborative frequency regulation according to the priority sequences and the total virtual reserve; periodically check the frequency regulation power deviation based on the real-time volatility of the DC bus voltage and the change rate of the SOC of the energy storage system, and adjust the dynamic economic weight according to the frequency regulation power deviation.

2. The method according to claim 1, characterized in that, The S1 includes: S11: Track the maximum power point of the photovoltaic array based on the incremental conductance method to obtain the maximum power that can be generated by the photovoltaic, and calculate the photovoltaic load reduction margin according to the formula ; where is the photovoltaic load reduction margin, is the maximum power that can be generated by the photovoltaic, is the output power of the photovoltaic array; S12: Calculate the charge and discharge power limits of the energy storage system according to the health state of the energy storage system using the following formula: ； where α and β are configuration parameters preset according to the energy storage type of the energy storage system, is the charge-discharge power limit of the energy storage system, is the rated charge-discharge power of the energy storage system, is the state of health of the energy storage system, is the reference value of the state of health of the energy storage system; S13: Calculate the energy storage discharge margin according to the charge and discharge power limits and the SOC of the energy storage system using the following formula: ； Among them, is the available discharge margin of the energy storage; is the state of charge of the energy storage system; is the maximum state of charge allowed for the energy storage system. S14: Identify the non-critical loads in the adjustable load through a load priority classification model to obtain the power of the non-critical loads, and calculate the load curtailment margin using the following formula: ； wherein, is the load shedding margin, is the power of the non-critical load, and γ is a preset load reduction coefficient.

3. The method according to claim 2, characterized in that, The calculation of the total virtual reserve includes: Perform weighted fusion of the PV load reduction margin, the energy storage discharge margin, and the load curtailment margin to obtain the total virtual reserve; the formula for the weighted fusion is: ； wherein, is the total virtual reserve, is the energy storage discharge efficiency coefficient of the energy storage system, is the load shedding efficiency coefficient.

4. The method according to claim 3, characterized in that, Before calculating the total virtual reserve, it also includes: S21: Monitor the temperature of the energy storage system. If the temperature exceeds the preset temperature threshold, reduce the energy storage discharge efficiency coefficient according to a linear relationship. S22: Correct the energy storage discharge efficiency coefficient according to the cycle times of the energy storage system using an exponential decay model. S23: Substitute the corrected energy storage discharge efficiency coefficient into the calculation of the total virtual reserve.

5. The method according to claim 2, characterized in that, The S3 includes: S31: Set a frequency regulation emergency weight according to the frequency regulation demand type. S32: Construct an economic cost function based on the real-time electricity price and the loss cost; the expression of the economic cost function is: ； Among them, is the dynamic economic weight calculated according to the real-time electricity price and the loss cost; S33: Input the total virtual reserve, the frequency regulation emergency weight, and the economic cost function into a preset fuzzy logic controller to generate the priority sequences through a preset rule base.

6. The method according to claim 2, characterized in that, The S4 includes: S41: Determine whether to preferentially call the PV load reduction margin according to the priority sequences; if so, calculate the PV power adjustment amount according to the frequency regulation demand and the PV load reduction margin, and generate a feed-forward reference power for the energy storage through the PV power adjustment amount and a feed-forward compensation coefficient; the formula for generating the feed-forward reference power for the energy storage is: ； Among them, is the feedforward reference power for energy storage; is the photovoltaic power adjustment amount, and satisfies ; is the feedforward compensation coefficient, , is the value of the DC bus capacitor, is the control delay; S42: Adjust the deviation of the DC bus voltage through a voltage outer loop PI controller to generate a feedback reference power for the energy storage; the formula for the feedback reference power for the energy storage is: ； Among them, is the energy storage feedback reference power, is at the DC bus voltage deviation at the moment, represents the accumulation of the DC bus voltage deviation over time, and are proportional-integral parameters set according to the dynamic response requirements of the energy storage system; S43: Superimpose the energy storage feedforward reference power and the energy storage feedback reference power to obtain the total energy storage reference power, and control the energy storage system to output power according to the total energy storage reference power.

7. According to the method according to any one of claims 1 to 6, characterized in that, The said S5 includes: S51: Allocate the photovoltaic load shedding margin, the energy storage discharge margin, and the load curtailment margin to the grid frequency regulation demand according to the priority sequence to generate the actual frequency regulation power. S52: Calculate the target deviation between the actual frequency regulation power and the target frequency regulation demand. If the ratio of the target deviation to the target frequency regulation demand is greater than a preset ratio, reallocate the dynamic economic weight based on a preset rule. S53: Correct the dynamic economic weight through the gradient descent algorithm based on the real-time volatility of the DC bus voltage and the state of charge change rate of the energy storage system.

Images