Renewable energy stable power generation method and device, equipment, storage medium

Abstract

The application provides a renewable energy stable power generation method and device, equipment and storage medium, and belongs to the field of energy generation. The method comprises the following steps: respectively extracting the operation data of a photovoltaic inverter and the charge state data of a power grid side in a preset time period to obtain power generation power parameters and power consumption power parameters; inputting the power generation power parameters and the power consumption power parameters into a source-grid matching degree evaluation model to obtain an evaluation result, the evaluation result being matching or not matching; if the evaluation result is not matching, determining a first control object and at least one second control object associated with the first control object according to the evaluation result; and generating a scheduling optimization instruction based on the predicted operation data of the control objects in a future preset time period, and sending the scheduling optimization instruction to a target object to perform stable power generation. The application can effectively improve the power generation stability of renewable energy, especially photovoltaic power generation.

Description

Technical Field

[0001] This application belongs to the field of energy power generation technology, and more specifically, relates to a method, apparatus, equipment, and storage medium for stable power generation of renewable energy. Background Technology

[0002] Photovoltaic (PV) power generation has become one of the core forms of renewable energy generation due to its clean, pollution-free nature and wide resource distribution. A PV power generation system includes PV arrays, PV inverters, and energy storage units. Among these, the PV inverter, as the core power conversion device between the PV array and the grid / load, undertakes key functions such as DC-AC power conversion, grid synchronization, and output power regulation, making it a crucial component for ensuring stable PV power output. However, existing PV inverter grid-connected regulation control often employs a step-response strategy, which easily triggers power output limiting or shutdown protection mechanisms, affecting the stability of power generation. Summary of the Invention

[0003] The purpose of this application is to provide a method, apparatus, equipment, and storage medium for stable power generation of renewable energy, so as to improve the stability of renewable energy power generation and extend the service life of inverters.

[0004] A first aspect of this application provides a method for stable renewable energy power generation, applied to a photovoltaic power generation system, the method comprising:

[0005] The power generation parameters are obtained by extracting features from the operating data of the photovoltaic inverter of the photovoltaic power generation system within the current preset time period, and the power consumption parameters are obtained by extracting features from the charge state data of the grid side within the current preset time period.

[0006] The power generation parameters and power consumption parameters are input into the source-grid matching degree evaluation model to obtain the evaluation result, which is either a match or a mismatch.

[0007] If the assessment result is a mismatch, the first regulatory object is determined based on the type label and severity level corresponding to the assessment result, and at least one second regulatory object associated with the first regulatory object is determined based on the security constraints and knowledge graph of the first regulatory object.

[0008] Based on the predicted operating data of the target object within a preset time period in the future, a scheduling optimization instruction is generated and sent to the target object to achieve stable power generation. The target object includes a first control object and at least one second control object.

[0009] A second aspect of this application provides a renewable energy stable power generation device applied to a photovoltaic power generation system, the device comprising:

[0010] The feature extraction module is used to extract features from the operating data of the photovoltaic inverter of the photovoltaic power generation system within the current preset time period to obtain the power generation parameters, and to extract features from the charge state data of the grid side within the current preset time period to obtain the power consumption parameters.

[0011] The evaluation module is used to input power generation parameters and power consumption parameters into the source-grid matching degree evaluation model to obtain the evaluation result, which is either a match or a mismatch.

[0012] The control object determination module is used to determine a first control object based on the type label and severity level corresponding to the assessment result if the assessment result is a mismatch, and to determine at least one second control object associated with the first control object based on the security constraints and knowledge graph of the first control object.

[0013] The optimization module is used to generate scheduling optimization instructions based on the predicted operating data of the target object within a preset time period in the future, and send the scheduling optimization instructions to the target object for stable power generation. The target object includes a first control object and at least one second control object.

[0014] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described method for stable renewable energy power generation.

[0015] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method for stable renewable energy power generation.

[0016] The beneficial effects of the renewable energy stable power generation method, apparatus, equipment, and storage medium provided in this application are as follows:

[0017] This embodiment first extracts the power generation and power consumption parameters corresponding to the photovoltaic inverter's operating data and the grid-side charge state data, respectively, and uses a source-grid matching degree evaluation model to accurately determine the matching status, changing the previous ambiguous judgment method. Secondly, when the evaluation result is a mismatch, this embodiment can determine the first control object based on type labels and severity levels, and use safety constraints and knowledge graphs to find the associated second control object. This comprehensively and systematically considers the coupling relationships between various parts of the system, thereby effectively avoiding the problems caused by step response strategies by regulating the first and second control objects, greatly improving the stability of photovoltaic power generation. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A schematic flowchart of a method for stable power generation of renewable energy provided in an embodiment of this application;

[0020] Figure 2 A structural block diagram of a renewable energy stable power generation device provided in an embodiment of this application;

[0021] Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0022] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0023] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0024] Please refer to Figure 1 , Figure 1 This is a schematic flowchart of a method for stable power generation of renewable energy provided in an embodiment of this application. It can be executed by an electronic device and is applied to a photovoltaic power generation system. It may include: S101~S104.

[0025] The photovoltaic power generation system includes a photovoltaic array, a photovoltaic inverter, an energy storage device, a monitoring and data acquisition unit, and a grid connection interface. The output of the photovoltaic array is connected to the DC input of the photovoltaic inverter, supplying DC power to the inverter. The AC output of the photovoltaic inverter is connected to the grid connection interface to achieve grid-connected output of the inverted power. The energy storage device is connected to the DC input of the photovoltaic inverter, enabling charge and discharge regulation to smooth fluctuations in power generation. The monitoring and data acquisition unit is connected to the photovoltaic array, photovoltaic inverter, energy storage device, and grid connection interface respectively, to collect operating data from each device.

[0026] S101: Extract features from the operating data of the photovoltaic inverter of the photovoltaic power generation system within the current preset time period to obtain the power generation parameters, and extract features from the charge state data of the grid side within the current preset time period to obtain the power consumption parameters.

[0027] In this embodiment, the current preset time period refers to a pre-set time period with a fixed time window, which is used to ensure that the data from the power generation side and the power consumption side are analyzed under the same time reference.

[0028] In one embodiment, the operating data of the photovoltaic inverter includes output power data and internal temperature data. The output power data includes, but is not limited to, the instantaneous active and reactive power of the AC side of the photovoltaic inverter dynamically changing over a preset time period, the rated power ratio, and the power fluctuation amplitude, which directly characterize the actual power generation capacity of the photovoltaic power generation system. The internal temperature data includes the temperature of the inverter's power conversion unit, the temperature of the heat dissipation unit, and the internal ambient temperature, reflecting the inverter's overload risk and health status, and is a key state quantity affecting the stability of power output.

[0029] This embodiment obtains feature quantities that can quantitatively characterize the real-time power generation capacity, output stability, and power change trend of a photovoltaic power generation system by performing feature extraction operations such as filtering and statistics on the operating data of the photovoltaic inverter.

[0030] In one embodiment, feature extraction is performed on the operating data of the photovoltaic inverter of the photovoltaic power generation system within a current preset time period to obtain power generation parameters, including:

[0031] The average power generation, maximum ramp rate, and power change rate of the photovoltaic inverter are obtained by feature extraction of the output power data based on a sliding window; the temperature change rate is obtained by feature extraction of the internal temperature data of the photovoltaic inverter.

[0032] The overload heating risk value of the photovoltaic inverter is obtained by coupling the power change rate and the temperature change rate.

[0033] The average power generation, maximum ramp rate, power change rate, temperature change rate, and overload heating risk value of the photovoltaic inverter are used as power generation parameters.

[0034] In this embodiment, the sliding window is a continuous time-series data processing time period constructed with a fixed window length and a fixed step size within the current preset time period. For example, if the preset time period is 10 minutes, the fixed window length of the sliding window can be 1 minute, and the fixed step size can be 30 seconds. Based on the sliding window, the output power data can be segmented, locally statistically analyzed, and dynamically characterized. While preserving the temporal correlation, it can smooth noise interference and quantify the power change pattern over a short period of time.

[0035] The average power generation represents a statistical characteristic obtained by performing an arithmetic or weighted average of the photovoltaic inverter output power data within the time interval corresponding to a single sliding window. It characterizes the overall power generation capacity of the photovoltaic power generation system within that time window. Here, output power can refer to active power. The maximum ramp rate represents the maximum increase or decrease in output power per unit time within a single sliding window. It is a quantitative indicator of the severity of output power fluctuations and characterizes the maximum change in photovoltaic power generation within a short period. The power change rate represents the rate of change of the photovoltaic inverter output power amplitude within the sliding window compared to the output power amplitude of the previous sliding window.

[0036] The rate of temperature change is a characteristic quantity obtained by differentiating the internal temperature data of key components such as the power conversion unit of a photovoltaic inverter. It is used to characterize the rate of temperature rise and fall of the internal temperature of the photovoltaic inverter over time, and directly reflects the rate of heat accumulation and heat dissipation efficiency of the photovoltaic inverter.

[0037] In this embodiment, the overload heating risk of the photovoltaic inverter is essentially the result of the interaction and synergistic aggravation of energy loss caused by power changes and heat accumulation caused by temperature changes, and cannot be quantified by a single parameter. The power change rate reflects the dynamic fluctuation intensity of the photovoltaic inverter's output power. Its magnitude directly reflects the conduction and switching losses of the power devices inside the inverter. The higher the output power of the photovoltaic inverter, the higher the instantaneous energy consumption of the power devices and the faster the heat generation rate. The temperature change rate reflects the dynamic evolution trend of the internal thermal field of the inverter. Its magnitude directly reflects the balance between heat dissipation and accumulation of the power devices. A positive temperature change rate and a larger value indicate that the heat accumulation rate exceeds the dissipation rate, and the device temperature will continue to rise. There is a significant coupling correlation between the two. That is, an increase in the power change rate will directly lead to an increase in the temperature change rate, and an abnormal increase in the temperature change rate will further aggravate the losses of the power devices, forming a vicious cycle of "power overload - heat accumulation - increased losses - sudden temperature rise", ultimately leading to overload heating failure. Therefore, it is necessary to quantify the synergistic effect of the two through coupling operations, and transform it into an overload heating risk value that can be directly used for risk assessment, so as to provide accurate parameter values ​​for subsequent source-network matching degree assessment.

[0038] As can be seen from the above, this embodiment extracts the core features of the photovoltaic inverter's output power data and internal temperature data, and performs coupled calculations of the power change rate and temperature change rate to quantify the inverter's overload heating risk. Furthermore, by using multi-dimensional features and overload heating risk values ​​together as power generation parameters, this embodiment can enrich the dimensions of parameter evaluation, improve the comprehensiveness and accuracy of parameter representation, provide effective data support for the subsequent source-grid matching degree evaluation model, and facilitate the final generation of stable power generation dispatch optimization instructions.

[0039] In this embodiment, the charge state data on the grid side includes power consumption data, including but not limited to the average power consumption on the grid side within the sliding window, the maximum ramp rate, and the power change rate. The calculation methods for the above parameters are the same as those for the average power generation, maximum ramp rate, and power change rate of a photovoltaic inverter, and will not be repeated here.

[0040] S102: Input the power generation parameters and power consumption parameters into the source-grid matching degree evaluation model to obtain the evaluation result, which is either a match or a mismatch.

[0041] In this embodiment, both the power generation parameter and the power consumption parameter are time-series feature quantities after standardized feature extraction. The two are time-synchronized and can be used to characterize the power generation capacity and dynamic fluctuation characteristics of the photovoltaic power generation side and the power demand and dynamic change pattern of the grid side within a preset time period. This provides the same source and synchronized basic data for the accurate evaluation of the source-grid matching degree assessment model.

[0042] In this embodiment, the source-grid matching evaluation model consists of a temporal feature encoding sub-model and a classification discriminant sub-model. The temporal feature encoding sub-model is a cascaded structure of a one-dimensional convolutional neural network and a long short-term memory network, used to map power generation parameters and power consumption parameters into source-grid matching feature vectors. The classification discriminant sub-model is a fully connected neural network and a Softmax classifier, used to evaluate the source-grid matching feature vectors to obtain evaluation results, and when the evaluation result is a mismatch, to determine the type label and severity level corresponding to the mismatch evaluation result.

[0043] In one embodiment, the type labels include overcapacity generation, undercapacity generation, and fluctuation matching, and the severity levels include mild, moderate, and severe; each type label corresponds to three severity levels.

[0044] Among them, the "excess power generation" type indicates that the power generation of the photovoltaic (PV) system consistently exceeds the power consumption on the grid side within a preset time period, resulting in surplus power. The "insufficient power generation" type indicates that the power generation of the PV system consistently falls short of the power consumption on the grid side within a preset time period, meaning the power generation of the PV system is far less than the power consumption on the grid side, and some grid-side electrical equipment lacks sufficient power to operate normally. The "fluctuation matching" type indicates that the time-series curves of power generation and power consumption exhibit high-frequency reverse fluctuations or phase misalignment, causing the difference between power generation and power consumption to fluctuate beyond the grid's safe tolerance range. This means that power generation may exceed power consumption at one moment, and then exceed it at the next, and so on, with repeated fluctuations.

[0045] The severity level is categorized as follows: Slight (meaning a match score greater than the first value), Moderate (meaning a match score less than or equal to the first value but greater than the second value), and Severe (meaning a match score less than or equal to the second value). The first value is greater than the second value, and both are within the range of (0,1). The first and second values ​​can be determined based on the historical power generation of the photovoltaic system and the power consumption on the grid side, with a match score greater than 0 and less than 1.

[0046] As can be seen from the above, this embodiment evaluates the matching results of power generation parameters and power consumption parameters by using a source-grid matching degree evaluation model. When the result is a mismatch, it can achieve fine-grained evaluation of the matching type and severity level, providing a reliable basis for subsequent differentiated equipment control.

[0047] S103: If the assessment result is a mismatch, then the first regulatory object is determined according to the type label and severity level corresponding to the assessment result, and at least one second regulatory object associated with the first regulatory object is determined based on the security constraints and knowledge graph of the first regulatory object.

[0048] In this embodiment, the first control object is obtained by querying the mapping table according to the type label and severity level corresponding to the evaluation result. The first control object can be each device in the photovoltaic power generation system, or it can be a component unit in each device, such as the maximum power point tracking component and LC filter unit of the photovoltaic inverter, the component support adjustment unit of the photovoltaic array, the energy storage unit of the energy storage device, etc.

[0049] In one embodiment, the knowledge graph construction process includes:

[0050] Each controllable object in the photovoltaic power generation system is used as a node in the knowledge graph;

[0051] The system acquires the attribute information and safety constraints of each control object in the photovoltaic power generation system, determines the electrical coupling relationship between each control object based on the attribute information and safety constraints, and uses the electrical coupling relationship between each control object as the edge of the knowledge graph.

[0052] In this embodiment, each device in the photovoltaic power generation system is coupled with another device. Adjusting the operating parameters of any device will generally affect the operating status of other devices. Therefore, after determining the first control object that needs to be adjusted, a second control object associated with the first control object must be determined. In this embodiment, the process of determining the second control object can be based on a knowledge graph.

[0053] First, the physical devices with power regulation capabilities in the photovoltaic power generation system are abstracted as nodes in a graph; these physical devices are the control objects. Second, the electrical coupling relationships between the various physical devices are identified. In this embodiment, the electrical coupling relationships between the control objects can be determined based on their attribute information and safety constraints.

[0054] The attribute information for each controlled object includes electrical attributes, control attributes, and structural attributes. Electrical attributes include, but are not limited to, voltage level, active / reactive power regulation capacity, electrical distance, and aging degree. Control attributes include, but are not limited to, regulation delay, regulation accuracy, and regulation loss coefficient. Structural attributes include, but are not limited to, equipment model and installation location.

[0055] Safety constraints refer to the rules that each regulated object must follow to avoid equipment damage and ensure the safe operation of the system during the execution of regulation actions and participation in power generation stability regulation. These constraints include upper limit constraints (maximum values ​​of voltage, current, power, or time delay), lower limit constraints (charge capacity), thermal stability constraints (temperature thresholds), and grid connection safety constraints (phase matching thresholds). Each regulated object corresponds to at least one safety constraint.

[0056] Electrical coupling relationships include power transmission correlation coefficients.

[0057] In one embodiment, determining the electrical coupling relationship between the various controlled objects based on their attribute information and safety constraints includes:

[0058] Each control object is mapped as a topology node, and a topology network structure is established based on the actual electrical wiring, signal connection and power flow path between each control object.

[0059] In the topological network structure, a power perturbation is injected into a single node, and the power response between the node and its neighboring nodes is calculated based on the attribute information. The power transfer gain and direction between the neighboring nodes are then calculated based on the power response.

[0060] Under the safety constraints of adjacent nodes, the electrical coupling relationship between each control object is determined based on the power transfer gain and direction between adjacent nodes.

[0061] In this embodiment, determining the electrical coupling relationship between each controlled object based on the power transfer gain and direction between adjacent nodes includes:

[0062] The power transmission correlation coefficient is calculated based on the power transfer gain and direction between adjacent nodes. For each pair of adjacent nodes, it is checked whether the power change of any node under the influence of the power transmission correlation coefficient will cause the other node to exceed the safety constraints; if it will not exceed the safety constraints, the power transmission correlation coefficient between the pair of adjacent nodes is taken as the electrical coupling relationship.

[0063] In this embodiment, the power transmission correlation coefficient is calculated based on the power transfer gain and direction between adjacent nodes, and can be obtained using the following formula:

[0064]

[0065] in, Indicates the power transmission correlation coefficient. Indicates power transfer gain. Represents a symbolic function. This indicates taking the minimum value. Represents a node i , j Maximum gain under safety constraints.

[0066] As can be seen from the above, this embodiment injects power perturbation into nodes and calculates the power transfer gain and direction between adjacent nodes, thereby obtaining the power transmission correlation coefficient (i.e., electrical coupling relationship), providing quantitative edge attributes for the knowledge graph and improving the accuracy of determining the second control object.

[0067] In one embodiment, determining at least one second regulatory object associated with the first regulatory object based on the security constraints and knowledge graph of the first regulatory object includes:

[0068] Starting with the first control object, traverse at least one candidate control object that is directly connected to the first control object through an edge in the established knowledge graph;

[0069] The constraint matching degree calculation and determination process is performed based on at least one candidate control object: the constraint matching degree between the safety constraint conditions of each candidate control object and the safety constraint conditions of the first control object is calculated respectively; the candidate control object with a constraint matching degree greater than the preset constraint matching degree threshold is taken as the second control object;

[0070] If a candidate control object directly connected by an edge does not meet the constraint matching degree requirement, then traverse other candidate control objects connected to the candidate control object and repeat the constraint matching degree calculation and judgment process until at least one candidate control object that meets the constraint matching degree requirement greater than the preset constraint matching degree threshold is determined or all nodes have been traversed.

[0071] In this embodiment, adjusting the first control object alone may exhibit instability over time. Therefore, this embodiment employs a structured traversal and constraint matching method within an established knowledge graph to determine the second control object associated with the first control object. The specific implementation process is as follows:

[0072] First, taking the currently determined first control object as the starting point of traversal, search for one or more candidate control objects in the knowledge graph that are directly connected to the first control object through an edge.

[0073] Secondly, for each directly connected candidate control object, constraint matching degree is calculated and determined. Specifically, the safety constraints of the candidate control object and the first control object are obtained respectively, and the constraint matching degree between them is calculated based on the compatibility of these constraints. This constraint matching degree is used to measure whether the candidate control object can maximize the stability of output power when working together without causing the first control object or itself to exceed the safety constraints. If the constraint matching degree of a candidate control object is greater than the preset constraint matching degree threshold, it is determined that it can be used as the second control object and participate in subsequent scheduling optimization together with the first control object. The preset constraint matching degree threshold can be set based on experience.

[0074] If no object satisfying the constraint matching degree requirement is found among the directly connected candidate control objects, the indirect connected candidate control objects of the directly connected candidate control objects are traversed based on the edges of the knowledge graph, and the constraint matching degree calculation and judgment process is executed again. This step can expand the search range layer by layer until at least one candidate control object satisfying the constraint matching degree greater than the threshold is found, or until all reachable nodes in the graph have been traversed.

[0075] In this embodiment, through the above-described traversal and matching process that expands layer by layer from near to far, a second control object that can form a synergistic effect with the first control object in terms of power balance and risk suppression can be dynamically identified within the scope of ensuring safety constraints, thereby achieving more reliable stable power generation control in the subsequent scheduling optimization stage.

[0076] S104: Generate scheduling optimization instructions based on the predicted operating data of the target object within a preset time period in the future, and send the scheduling optimization instructions to the target object for stable power generation. The target object includes a first control object and at least one second control object.

[0077] In this embodiment, in order to achieve matching between photovoltaic power generation and grid load within a preset future time period and maintain the stability of the power generation process, it is necessary to formulate a reasonable scheduling optimization scheme for the first control object and at least one second control object participating in the coordinated control based on the predicted operating data of the target object within the preset future time period, and issue executable control commands. The specific process is as follows:

[0078] The system acquires predicted operational data for the first and second regulated objects within a predetermined future time period. This predicted operational data can be obtained using a long short-term memory network based on historical data, and the predicted operational data for each regulated object are kept synchronized on the timeline.

[0079] Based on the predicted operating data of the first and second controlled objects within a preset future time period, an optimization algorithm is used to generate a power adjustment-based scheduling optimization scheme, aiming to reduce energy storage device losses and lower the probability of the first and second controlled objects triggering their respective safety constraints during operation. Finally, after generating scheduling optimization instructions based on the scheduling optimization scheme, the feasibility of the instructions is verified, and instructions that meet the feasibility verification are sent to the first and second controlled objects to achieve stable power generation of the photovoltaic power generation system.

[0080] In this embodiment, a gradual trajectory planning method is used during power adjustment to smoothly design the power adjustment process for the target object, ensuring that the power change rate of any adjusted object does not exceed the upper limit of the allowable power change rate corresponding to its current temperature. This method avoids step-like impacts, thereby reducing the risk of triggering protective measures on the photovoltaic inverter and other power electronic equipment.

[0081] As can be seen from the above, this embodiment first extracts the power generation parameters and power consumption parameters corresponding to the photovoltaic inverter operation data and grid-side charge state data, respectively, and accurately judges the matching status using a source-grid matching degree evaluation model, changing the previous fuzzy judgment method. Secondly, when the evaluation result is a mismatch, this embodiment can determine the first control object based on type labels and severity levels, and find the associated second control object using safety constraints and knowledge graphs. This comprehensively and systematically considers the coupling relationships between various parts of the system, thereby effectively avoiding the problems caused by step response strategies by regulating the first and second control objects, greatly improving the stability of photovoltaic power generation.

[0082] In one embodiment of this application, the overload heating risk value of a photovoltaic inverter is obtained by coupling the power change rate and the temperature change rate, including:

[0083] The power change rate is mapped to a power excitation feature vector, and the temperature change rate is mapped to a thermal response feature vector. Tensor inner product operation is performed on the power excitation feature vector and the thermal response feature vector to obtain the power-thermal coupling feature value.

[0084] Based on the preset inverter health benchmark value, the power-thermal coupling characteristic value is calculated to obtain the coupling residual characteristic value, and the overload heating risk value of the inverter is obtained based on the coupling residual characteristic value.

[0085] In this embodiment, the process of calculating the power-thermal coupling characteristic value can be understood as follows:

[0086] A two-dimensional coupled feature space based on time and parameter dimensions is constructed. The horizontal axis of this feature space represents the time series nodes within the current preset time period (corresponding one-to-one with the time nodes from which the sliding window extracts data), and the vertical axis represents the two-dimensional parameter dimension of "power excitation-thermal response". The extracted power change rate ΔP and temperature change rate ΔT are mapped to the power excitation feature vector P and the thermal response feature vector T within this feature space, respectively. The power excitation feature vector P is composed of the ΔP values ​​at each time node arranged in chronological order, and the thermal response feature vector T is composed of the ΔT values ​​at each time node arranged in chronological order. The dimension of the vectors is consistent with the number of sliding windows within the current preset time period.

[0087] Based on the constructed two-dimensional coupling feature space, a tensor inner product operation is performed on the power excitation feature vector P and the thermal response feature vector T to obtain the power-thermal coupling feature value K. This value is a scalar value, and its magnitude directly represents the cooperative coupling strength between the power change rate and the temperature change rate of the photovoltaic inverter within a preset time period. In this embodiment, the preset inverter health benchmark value is a value obtained by calibrating experimental data or simulation data of the photovoltaic inverter under health conditions. The power-thermal coupling feature value K is calculated in conjunction with the preset inverter health benchmark value. The difference between them can be used to obtain the coupling residual characteristic value ΔK.

[0088] In this embodiment, the overload heating risk value of the inverter is obtained based on the coupling residual feature value. That is, the overload heating risk value is calculated using the overload risk discrimination kernel function on the coupling residual feature value. The overload risk discrimination kernel function is:

[0089] ;

[0090] This kernel function uses sigmoid Function and Risk Value Scaling Factor C The combined approach can enable sigmoidThe function's output range (0,1) is mapped to a preset risk assessment interval (0,10). In this embodiment, by inputting the coupling residual characteristic value into the overload risk discrimination kernel function, the overload heating risk value of the photovoltaic inverter can be obtained. F= .

[0091] As can be seen from the above, this embodiment can characterize the dynamic coupling relationship between the power change rate and the temperature change rate by mapping them to feature vectors and performing tensor inner product operations. Combined with the inverter health benchmark value for residual calculation, it can effectively eliminate interference under normal operating conditions, accurately identify abnormal overload heating trends, improve the reliability and sensitivity of risk assessment, and provide more robust feature inputs for subsequent source-grid matching assessment.

[0092] Corresponding to the renewable energy stable power generation method in the above embodiments, Figure 2 This is a structural block diagram of a renewable energy stable power generation device according to an embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. Reference Figure 2 The renewable energy stable power generation device 20 is applied to a photovoltaic power generation system and includes: a feature extraction module 21, an evaluation module 22, a control object determination module 23, and an optimization module 24.

[0093] Among them, the feature extraction module 21 is used to extract features from the operating data of the photovoltaic inverter of the photovoltaic power generation system within the current preset time period to obtain the power generation parameters, and to extract features from the charge state data of the grid side within the current preset time period to obtain the power consumption parameters.

[0094] Evaluation module 22 is used to input power generation parameters and power consumption parameters into the source-grid matching degree evaluation model to obtain evaluation results, which are either matching or mismatched.

[0095] The control object determination module 23 is used to determine a first control object based on the type label and severity level corresponding to the assessment result if the assessment result is a mismatch, and to determine at least one second control object associated with the first control object based on the security constraints and knowledge graph of the first control object.

[0096] The optimization module 24 is used to generate scheduling optimization instructions based on the predicted operating data of the target object within a preset time period in the future, and send the scheduling optimization instructions to the target object for stable power generation. The target object includes a first control object and at least one second control object.

[0097] In one embodiment of this application, the operating data of the photovoltaic inverter includes output power data and internal temperature data of the photovoltaic inverter; the feature extraction module 21 is specifically used for:

[0098] The average power generation, maximum ramp rate, and power change rate of the photovoltaic inverter are obtained by feature extraction of the output power data based on a sliding window; the temperature change rate is obtained by feature extraction of the internal temperature data of the photovoltaic inverter.

[0099] The overload heating risk value of the photovoltaic inverter is obtained by coupling the power change rate and the temperature change rate.

[0100] The average power generation, maximum ramp rate, power change rate, temperature change rate, and overload heating risk value of the photovoltaic inverter are used as power generation parameters.

[0101] In one embodiment of this application, when the feature extraction module 21 couples the power change rate and temperature change rate to obtain the overload heating risk value of the photovoltaic inverter, it is specifically used for:

[0102] The power change rate is mapped to a power excitation feature vector, and the temperature change rate is mapped to a thermal response feature vector. Tensor inner product operation is performed on the power excitation feature vector and the thermal response feature vector to obtain the power-thermal coupling feature value.

[0103] Based on the preset inverter health benchmark value, the power-thermal coupling characteristic value is calculated to obtain the coupling residual characteristic value, and the overload heating risk value of the inverter is obtained based on the coupling residual characteristic value.

[0104] In one embodiment of this application, the renewable energy stable power generation device 20 further includes a knowledge graph construction module, which, when building the knowledge graph, is specifically used for:

[0105] Each controllable object in the photovoltaic power generation system is used as a node in the knowledge graph;

[0106] The system acquires the attribute information and safety constraints of each control object in the photovoltaic power generation system, determines the electrical coupling relationship between each control object based on the attribute information and safety constraints, and uses the electrical coupling relationship between each control object as the edge of the knowledge graph.

[0107] In one embodiment of this application, when determining the electrical coupling relationship between various control objects based on the attribute information and security constraints of each control object, the knowledge graph construction module is specifically used for:

[0108] Each control object is mapped as a topology node, and a topology network structure is established based on the actual electrical wiring, signal connection and power flow path between each control object.

[0109] In the topological network structure, a power perturbation is injected into a single node, and the power response between the node and its neighboring nodes is calculated based on the attribute information. The power transfer gain and direction between the neighboring nodes are then calculated based on the power response.

[0110] Under the safety constraints of adjacent nodes, the electrical coupling relationship between each control object is determined based on the power transfer gain and direction between adjacent nodes.

[0111] In one embodiment of this application, when the regulation object determination module 23 determines at least one second regulation object associated with the first regulation object based on the security constraints and knowledge graph of the first regulation object, it is specifically used for:

[0112] Starting with the first control object, traverse at least one candidate control object that is directly connected to the first control object through an edge in the established knowledge graph;

[0113] The constraint matching degree calculation and determination process is performed based on at least one candidate control object: the constraint matching degree between the safety constraint conditions of each candidate control object and the safety constraint conditions of the first control object is calculated respectively; the candidate control object with a constraint matching degree greater than the preset constraint matching degree threshold is taken as the second control object;

[0114] If a candidate control object directly connected by an edge does not meet the constraint matching degree requirement, then traverse other candidate control objects connected to the candidate control object and repeat the constraint matching degree calculation and judgment process until at least one candidate control object that meets the constraint matching degree requirement greater than the preset constraint matching degree threshold is determined or all nodes have been traversed.

[0115] In one embodiment of this application, the type labels corresponding to the assessment results include power generation surplus type, power generation shortage type, and fluctuation matching type, and the severity levels include slight, moderate, and severe; each type label corresponds to three severity levels.

[0116] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of the modules in the aforementioned device embodiments, for example... Figure 2 The functions of the feature extraction module 21, evaluation module 22, regulation object determination module 23, and optimization module 24 are shown.

[0117] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), but it may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0118] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0119] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store device type information.

[0120] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation methods described in the renewable energy stable power generation method provided in the embodiments of this application, or they can execute the implementation methods of the electronic devices described in the embodiments of this application, which will not be repeated here.

[0121] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0122] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0123] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0124] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0125] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or modules, or it may be an electrical, mechanical, or other form of connection.

[0126] The modules described as separate components may or may not be physically separate. Similarly, the components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0127] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0128] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for stable power generation from renewable energy sources, characterized in that, Applications in photovoltaic power generation systems include: The power generation parameters are obtained by extracting features from the operating data of the photovoltaic inverter of the photovoltaic power generation system within the current preset time period, and the power consumption parameters are obtained by extracting features from the charge state data of the grid side within the current preset time period; the operating data of the photovoltaic inverter includes output power data and internal temperature data of the photovoltaic inverter. The power generation parameters and the power consumption parameters are input into the source-grid matching degree evaluation model to obtain the evaluation result, which is either a match or a mismatch. If the assessment result is a mismatch, then a first regulatory object is determined based on the type label and severity level corresponding to the assessment result, and at least one second regulatory object associated with the first regulatory object is determined based on the security constraints and knowledge graph of the first regulatory object. Based on the predicted operating data of the target object within a preset time period in the future, a scheduling optimization instruction is generated and sent to the target object to achieve stable power generation. The target object includes a first control object and at least one second control object. The step of extracting features from the operating data of the photovoltaic inverter of the photovoltaic power generation system within the current preset time period to obtain the power generation parameters includes: The average power generation, maximum ramp rate, and power change rate of the photovoltaic inverter are obtained by feature extraction of the output power data based on a sliding window; the temperature change rate is obtained by feature extraction of the internal temperature data of the photovoltaic inverter. The overload heating risk value of the photovoltaic inverter is obtained by coupling the power change rate and the temperature change rate. The average power generation, maximum ramp rate, power change rate, temperature change rate, and overload heating risk value of the photovoltaic inverter are used as power generation parameters. The process of establishing the knowledge graph includes: Each controllable object in the photovoltaic power generation system is used as a node in the knowledge graph; Obtain the attribute information and safety constraints of each control object in the photovoltaic power generation system, determine the electrical coupling relationship between each control object based on the attribute information and safety constraints of each control object, and use the electrical coupling relationship between each control object as the edge of the knowledge graph; The step of determining at least one second regulatory object associated with the first regulatory object based on the security constraints and knowledge graph of the first regulatory object includes: Starting with the first control object, traverse at least one candidate control object that is directly connected to the first control object through the edge in the established knowledge graph; Based on the at least one candidate control object, a constraint matching degree calculation and determination process is performed: the constraint matching degree between the safety constraint conditions of each candidate control object and the safety constraint conditions of the first control object is calculated respectively; the candidate control object with the constraint matching degree greater than the preset constraint matching degree threshold is taken as the second control object; If the candidate control object directly connected by the edge does not meet the constraint matching degree requirement, then traverse other candidate control objects connected to the candidate control object and repeat the constraint matching degree calculation and judgment process until at least one candidate control object that meets the constraint matching degree requirement greater than the preset constraint matching degree threshold is determined or all nodes have been traversed.

2. The method for stable power generation of renewable energy as described in claim 1, characterized in that, The process of coupling the power change rate and the temperature change rate to obtain the overload heating risk value of the photovoltaic inverter includes: The power change rate is mapped to a power excitation feature vector, and the temperature change rate is mapped to a thermal response feature vector. Tensor inner product operation is performed on the power excitation feature vector and the thermal response feature vector to obtain power-thermal coupling feature values. Based on the preset inverter health benchmark value, the power-thermal coupling characteristic value is calculated to obtain the coupling residual characteristic value, and the overload heating risk value of the inverter is obtained based on the coupling residual characteristic value.

3. The method for stable power generation of renewable energy as described in claim 1, characterized in that, Determining the electrical coupling relationship between the various control objects based on their attribute information and safety constraints includes: Each control object is mapped as a topology node, and a topology network structure is established based on the actual electrical wiring, signal connection and power flow path between each control object. In the aforementioned topology, a power perturbation is injected into a single node, and the power response between the node and its neighboring nodes is calculated based on the attribute information. The power transfer gain and direction between the neighboring nodes are then calculated based on the power response. Under the safety constraints of adjacent nodes, the electrical coupling relationship between each control object is determined based on the power transfer gain and direction between the adjacent nodes.

4. The method for stable power generation of renewable energy as described in claim 1, characterized in that, The assessment results correspond to three types of labels: overcapacity, undercapacity, and fluctuation matching. The severity levels are mild, moderate, and severe. Each type label corresponds to three severity levels.

5. A renewable energy stable power generation device, characterized in that, The device is used in photovoltaic power generation systems and includes: The feature extraction module is used to extract features from the operating data of the photovoltaic inverter of the photovoltaic power generation system within the current preset time period to obtain the power generation parameters, and to extract features from the charge state data of the grid side within the current preset time period to obtain the power consumption parameters; the operating data of the photovoltaic inverter includes output power data and internal temperature data of the photovoltaic inverter; The evaluation module is used to input the power generation parameters and the power consumption parameters into the source-grid matching degree evaluation model to obtain the evaluation result, which is either a match or a mismatch. The control object determination module is used to determine a first control object based on the type label and severity level corresponding to the assessment result if the assessment result is a mismatch, and to determine at least one second control object associated with the first control object based on the security constraints and knowledge graph of the first control object. An optimization module is used to generate a scheduling optimization instruction based on the predicted operating data of the target object within a preset time period in the future, and send the scheduling optimization instruction to the target object to achieve stable power generation. The target object includes a first control object and at least one second control object. The feature extraction module is specifically used for: The average power generation, maximum ramp rate, and power change rate of the photovoltaic inverter are obtained by feature extraction of the output power data based on a sliding window; the temperature change rate is obtained by feature extraction of the internal temperature data of the photovoltaic inverter. The overload heating risk value of the photovoltaic inverter is obtained by coupling the power change rate and the temperature change rate. The average power generation, maximum ramp rate, power change rate, temperature change rate, and overload heating risk value of the photovoltaic inverter are used as power generation parameters. The renewable energy stable power generation device also includes a knowledge graph construction module, which, when building the knowledge graph, is specifically used for: Each controllable object in the photovoltaic power generation system is used as a node in the knowledge graph; Obtain the attribute information and safety constraints of each control object in the photovoltaic power generation system, determine the electrical coupling relationship between each control object based on the attribute information and safety constraints of each control object, and use the electrical coupling relationship between each control object as the edge of the knowledge graph; The module for determining the control target is specifically used for: Starting with the first control object, traverse at least one candidate control object that is directly connected to the first control object through the edge in the established knowledge graph; Based on the at least one candidate control object, a constraint matching degree calculation and determination process is performed: the constraint matching degree between the safety constraint conditions of each candidate control object and the safety constraint conditions of the first control object is calculated respectively; the candidate control object with the constraint matching degree greater than the preset constraint matching degree threshold is taken as the second control object; If the candidate control object directly connected by the edge does not meet the constraint matching degree requirement, then traverse other candidate control objects connected to the candidate control object and repeat the constraint matching degree calculation and judgment process until at least one candidate control object that meets the constraint matching degree requirement greater than the preset constraint matching degree threshold is determined or all nodes have been traversed.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes a computer program, it implements the steps of the method as claimed in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that, When a computer program is executed by a processor, it implements the steps of the method as claimed in any one of claims 1 to 4.

Images