Photovoltaic energy storage charging and discharging behavior monitoring system and method

Abstract

The application relates to the field of charge-discharge behavior monitoring, and discloses a photovoltaic energy storage charge-discharge behavior monitoring system and method, which comprises the following steps: continuously collecting energy operation data of an energy storage unit, a photovoltaic power generation side, a grid-connected side and a load side to construct a continuous time sequence energy operation model; in the continuous time sequence energy operation model, based on the change characteristics of the energy storage power direction and the dynamic evolution law of the state of charge, a charge-discharge behavior fragment set is formed; a multi-dimensional behavior correlation network structure of the charge-discharge behavior fragment is constructed, and an energy path mapping relationship is established; an effective energy displacement criterion is introduced, the adjustment contribution degree evaluation of the correlated charge-discharge behavior fragment is performed, and it is determined that the energy storage operation is in an adjustment contribution degree failure state; based on a behavior triggering rule, the adjustment contribution degree failure state is discretized into a structured operation behavior event set, and energy storage operation strategy adjustment prompt information is generated; and the application has the advantages of improving the adjustment accuracy and reducing the energy storage loss.

Description

Technical Field

[0001] This invention relates to the field of charging and discharging behavior monitoring, specifically to a photovoltaic energy storage charging and discharging behavior monitoring system and method. Background Technology

[0002] With the widespread application of photovoltaic power generation systems and energy storage devices in new energy power plants and distributed energy scenarios, existing technologies typically monitor parameters such as voltage, current, power, and state of charge of energy storage units through battery management systems or energy management systems to identify their charging, discharging, or standby states and ensure operational safety. However, in actual operation, energy storage systems need to work in conjunction with photovoltaic power generation units, grid-connected interfaces, and local loads. Under conditions of light fluctuations, load changes, or strategy switching, energy storage may charge and then quickly discharge within a short period of time, and the corresponding electrical energy is ultimately utilized through the original grid connection or load absorption path, without forming an effective time shift or energy transfer. Although such charging and discharging processes meet operational constraints in terms of state parameters, from the perspective of the overall system, they only constitute energy transfer, increasing unnecessary energy loss and battery life loss. Existing charging and discharging monitoring technologies mainly focus on whether charging and discharging states occur, lacking the ability to comprehensively judge the charging and discharging behavior in terms of time correlation, energy destination, and system benefits. It is difficult to identify the aforementioned inefficient charging and discharging behaviors that do not form an effective energy shift, thus failing to provide sufficient monitoring basis for system operation analysis and maintenance management. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a photovoltaic energy storage charging and discharging behavior monitoring system and method, which has the advantages of improving regulation accuracy and reducing energy storage losses, thus solving the problems mentioned in the background technology.

[0004] To achieve the aforementioned goals of improving regulation accuracy and reducing energy storage losses, this invention provides the following technical solution: a method for monitoring the charging and discharging behavior of photovoltaic energy storage, comprising the following steps: By synchronously collecting energy operation data from energy storage units, photovoltaic power generation side, grid connection side and load side, and performing reconstruction processing on multi-sampling period data based on a unified time reference, a continuous time-series energy operation model of the overall energy flow state is constructed. In the continuous time-series energy operation model, based on the characteristics of energy storage power direction change and the dynamic evolution law of state of charge, the execution behavior of the energy storage operation process is structurally decomposed to form a set of charging and discharging behavior segments with time boundaries and energy attribute identifiers. Based on the set of charging and discharging behavior segments, a multi-dimensional behavior association network structure of charging and discharging behavior segments is constructed. By analyzing the change characteristics of grid-connected power and load power during charging and discharging, an energy path mapping relationship between charging and discharging behavior segments and energy flow direction is established. Based on the energy path mapping relationship, an effective energy displacement criterion is introduced to evaluate the regulation contribution of associated charging and discharging behavior segments. When the charging and discharging behavior is highly coupled in the time dimension and no effective energy regulation displacement is formed, the energy storage operation is determined to be in a state of regulation contribution failure. Based on behavior triggering rules, the failure state of adjustment contribution is discretized into a set of structured operation behavior events. Combined with the characteristics of energy storage operation cycle, the execution cycle of operation behavior events is matched and analyzed to generate energy storage operation strategy adjustment prompt information.

[0005] Preferably, the process of constructing a continuous time-series energy operation model of the overall energy flow state is as follows: The charging and discharging power data, state of charge data, and battery operating status data of the energy storage unit are collected in real time. At the same time, the output power data of the photovoltaic power generation side, the switching power data of the grid-connected side, and the energy consumption power data of the load side are acquired synchronously. Time alignment is performed on data with different sampling frequencies based on a unified time reference; By constructing a time index mapping relationship, missing sampled data is interpolated and repaired, and abnormal fluctuation data is denoised and smoothed to form a data sequence with a unified time granularity. Based on the energy conservation principle, the energy exchange process of the energy storage side, power generation side, grid connection side and load side is checked for consistency of flow direction, and a multi-node energy flow coupling relationship matrix is ​​constructed. Based on the coupling relationship matrix, time-series fusion processing is performed on the energy exchange data of multiple nodes to generate a continuous time-series energy operation model that reflects the overall energy input, storage, output and consumption relationship of the system.

[0006] Preferably, the process of forming a set of charging and discharging behavior segments with time boundaries and energy attribute identifiers is as follows: In the continuous time-series energy operation model, identify the inflection points of positive and negative changes in energy storage power, and divide the energy storage operation stage according to the continuous interval of power direction; By combining the rate of change of state of charge, power stability, and cumulative energy change, boundary corrections are performed on the energy storage operation phase to form an operating range that is time-continuous and physically consistent. Extract the charging / discharging type identifier, duration parameter, energy change amplitude parameter, and power change characteristic parameter for each operating range; Each operating interval is divided into a set of charging and discharging behavior segments with clear time boundaries and energy attribute identifiers.

[0007] Preferably, the process of constructing a multidimensional behavioral association network structure for charging and discharging behavior segments is as follows: The charging and discharging behavior segments are sequentially arranged according to time order to establish the temporal adjacency relationship between the behavior segments; Based on the direction and amount of energy change of behavioral segments, establish the correlation of energy transfer of behavioral segments and construct the energy inheritance mapping relationship between behavioral segments; By combining the grid-connected power change trend and load power response characteristics corresponding to the behavior segments, the correlation degree of energy regulation coordination between behavior segments is calculated. Based on temporal adjacency, energy inheritance mapping, and regulatory coordination, a multidimensional behavioral association network structure for charging and discharging behavior segments is constructed.

[0008] Preferably, the process of establishing the energy path mapping relationship between charging / discharging behavior segments and energy flow direction is as follows: In the multidimensional behavior association network structure, the combination relationship between each charging behavior segment and the corresponding discharging behavior segment is extracted; Based on the combination relationship of discharge behavior segments, the grid-connected power change, load power change and photovoltaic output change during each behavior segment are analyzed to identify the energy source node of energy storage and the target node of energy release. Generate an energy flow sequence along the energy input / output path from the source node to the target node; Each charging and discharging behavior segment is mapped to its corresponding energy flow sequence, forming an energy path mapping relationship between the charging and discharging behavior segments and the energy flow direction.

[0009] Preferably, the process for evaluating the moderating contribution of correlated charge-discharge behavior segments is as follows: Based on the energy path mapping relationship, the energy input path and energy output path corresponding to each charging and discharging behavior segment are extracted; For each path, energy time migration, energy spatial transfer, and energy regulation response contribution are constructed as evaluation factors for effective energy displacement criteria. Based on the evaluation factors, the regulation effect of energy storage on system load changes, grid-connected power balance and photovoltaic fluctuation suppression in the energy input and output stages is analyzed, and the regulation contribution of each charging and discharging behavior segment to the system operation stability is calculated. Based on the magnitude of the regulatory contribution, the charging and discharging behavior segments are classified into value levels to form a regulatory contribution evaluation result.

[0010] Preferably, the process for determining that energy storage operation is in a state of regulatory contribution failure is as follows: Based on the evaluation results of the regulation contribution, charging and discharging behavior segments with regulation contribution below a preset threshold are selected. For the selected behavioral segments, analyze the correspondence between energy input and output of charging and discharging behaviors within a preset time window, and determine the effect of the correspondence on the improvement of the load curve. When the correspondence between energy input and output does not significantly improve the system's operating state, further analysis should be conducted on the spatial consistency between the energy input path and the energy output path. If the energy input path and the output path point to the same node, the energy spatial migration effect is deemed insufficient. When energy storage behavior exhibits both low regulation contribution and energy circulation within the time window without significantly improving load and grid-connected power fluctuations, and the energy spatial migration effect is insufficient, it is determined that the energy storage operation is in a state of regulation contribution failure.

[0011] Preferably, the process of discretizing the failure state of adjustment contribution into a set of structured operational behavior events based on behavior triggering rules is as follows: Based on the combined characteristics of behavioral segments corresponding to the failure state of the adjustment contribution, the time start and end nodes, energy change amplitude and cycle number parameters of the failure behavior are extracted. Clustering is performed on consecutive failure behaviors based on behavior triggering rules to generate runtime behavior events with unique identifiers; Each operational behavior event is assigned a behavior type label, a failure level label, and an energy cycle characteristic label to form a structured set of operational behavior events.

[0012] Preferably, the process of generating energy storage operation strategy adjustment prompts is as follows: Extract the operational characteristics of the energy storage system under daily, load, and photovoltaic power generation cycles; By performing time-aligned analysis of the structured set of operational behavior events and the characteristics of each cycle, the recurrence pattern of inefficient behavior in different operational cycles can be identified. The operational risk level is assessed based on the frequency of inefficient behavior, energy loss, and the degree of impact on system stability. Based on risk level analysis, the adjustment needs of the energy storage system are analyzed, and information prompts for adjusting the energy storage operation strategy are generated.

[0013] A photovoltaic energy storage charging and discharging behavior monitoring system includes: Energy modeling module: Synchronously collects data from energy storage, photovoltaics, grid connection and load, and constructs a continuous time-series energy operation model of the overall energy flow; Behavior decomposition module: Based on the power change and state of charge evolution laws, the energy storage operation process is decomposed into a set of charging and discharging behavior segments with time boundaries and energy attribute labels; Network association module: Constructs a multi-dimensional behavior association network based on behavior segments, and forms an energy path mapping relationship between charging and discharging behavior segments and energy flow direction; Contribution evaluation module: Utilizes energy path mapping relationship and effective displacement criterion to analyze the moderating contribution of behavioral segments and identify the failure state of moderating contribution; Strategy prompt module: Discretizes the adjustment failure state into structured operation events, and generates energy storage operation strategy adjustment prompt information in combination with the operation cycle.

[0014] Compared with the prior art, the present invention provides a photovoltaic energy storage charging and discharging behavior monitoring system and method, which has the following beneficial effects: This invention achieves refined decomposition and structured characterization of energy storage charging and discharging behavior by synchronously acquiring and continuously time-series modeling multi-source energy data from energy storage units, photovoltaic power generation side, grid connection side, and load side. Based on multi-dimensional behavior correlation networks and energy path mapping, it can accurately identify the energy input-output relationship of each charging and discharging behavior segment in time and space, quantify its contribution to system load balance, grid-connected power stability, and photovoltaic fluctuation suppression, and distinguish between effective energy migration behavior and inefficient energy cycling behavior, thereby discovering behaviors that may cause energy waste and battery life loss. Through event-based analysis of the failure state of adjustment contribution, it can statistically identify periodic inefficient patterns of energy storage behavior and generate targeted operation strategy adjustment prompts to guide system optimization scheduling. Compared with traditional methods that only monitor charging and discharging status, this invention achieves innovative management from simple charging and discharging status monitoring to overall system energy regulation optimization, significantly improving system regulation accuracy and reducing energy storage losses. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the method of the present invention; Figure 2 This is a schematic diagram of the structure of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example 1: Please refer to Figure 1 As shown in the figure, a photovoltaic energy storage charging and discharging behavior monitoring method according to an embodiment of the present invention includes the following steps: S1: By synchronously collecting energy operation data from energy storage units, photovoltaic power generation side, grid connection side and load side, and performing reconstruction processing on multi-sampling period data based on a unified time reference, a continuous time-series energy operation model of the overall energy flow state is constructed.

[0018] The process of constructing a continuous time-series energy operation model of the overall energy flow state in S1 is as follows: The charging and discharging power data, state of charge data, and battery operating status data of the energy storage unit are collected in real time. At the same time, the output power data of the photovoltaic power generation side, the switching power data of the grid-connected side, and the energy consumption power data of the load side are acquired synchronously. By installing high-precision current, voltage, and temperature sensors inside the energy storage system, the power changes, state of charge, and health status of the energy storage unit are monitored in real time. At the same time, corresponding power metering devices are deployed at the photovoltaic array port, grid connection interface, and load measurement point to ensure that data from each node can be collected synchronously with a time resolution of milliseconds or seconds. The data is uploaded to the data processing module in real time through a unified communication bus or acquisition gateway to form a preliminary multi-source heterogeneous energy data set.

[0019] Time alignment is performed on data with different sampling frequencies based on a unified time reference; By selecting a unified system time reference, such as a GPS timestamp or a local high-precision clock, as the alignment benchmark, the data streams of energy storage units, photovoltaic power generation, grid connection side and load side are processed with unified time stamps. For data with different sampling frequencies, up or down resampling can be used to map all data to a unified time grid, thereby ensuring that the data of each node can be directly corresponded on the same time scale in subsequent energy flow analysis, realizing the synchronized management of cross-system and multi-node data.

[0020] By constructing a time index mapping relationship, missing sampled data is interpolated and repaired, and abnormal fluctuation data is denoised and smoothed to form a data sequence with a unified time granularity. By traversing a unified time series, missing data points caused by acquisition delays or communication anomalies are identified. Linear interpolation, spline interpolation, or dynamic prediction algorithms based on historical sequences are used to fill in the data. For abnormal fluctuation points caused by environmental disturbances or measurement noise, methods such as moving average filtering, exponential smoothing, or wavelet denoising are used for smoothing, thereby generating a continuous and stable time series, providing a reliable basis for energy flow analysis.

[0021] Based on the energy conservation principle, the energy exchange process of the energy storage side, power generation side, grid connection side and load side is checked for consistency of flow direction, and a multi-node energy flow coupling relationship matrix is ​​constructed. By substituting the input power, output power, and power consumption data of each node at the same time scale into the energy conservation formula, the total energy balance error of the system is calculated. When local inconsistencies or abnormal deviations are found, corrections are made by adjusting the measurement error weights or combining historical data. Finally, a matrix reflecting the energy flow coupling relationship between each node is constructed. Each element in the matrix represents the quantity and direction of energy flowing from one node to another within a specific time period, thereby realizing a quantitative description of the energy transfer relationship within the system.

[0022] Based on the coupling relationship matrix, time-series fusion processing is performed on the energy exchange data of multiple nodes to generate a continuous time-series energy operation model that reflects the overall energy input, storage, output and consumption relationship of the system. By combining the coupling relationship matrix with unified time series data, matrix operations and cumulative analysis are performed in chronological order to form a continuous mapping of energy flow across multiple nodes. At the same time, energy loss rate, energy storage efficiency, and power generation fluctuation factors can be introduced into the model, so that the model not only describes the energy flow state, but also accurately reflects the dynamic relationship of energy input, storage, output and consumption in actual system operation. Finally, a continuous time series model that can be used to monitor, optimize and predict the overall energy operation behavior of photovoltaic energy storage systems is generated.

[0023] S2: In the continuous time-series energy operation model, based on the characteristics of energy storage power direction change and the dynamic evolution law of charge state, the execution behavior of energy storage operation process is structurally decomposed to form a set of charging and discharging behavior segments with time boundaries and energy attribute identifiers.

[0024] The process of forming a set of charging and discharging behavior segments with time boundaries and energy attribute identifiers in S2 is as follows: In the continuous time-series energy operation model, identify the inflection points of positive and negative changes in energy storage power, and divide the energy storage operation stage according to the continuous interval of power direction; By performing first-order difference or sliding slope analysis on the continuous power time series of energy storage units, the inflection points where power changes from positive discharge to negative charging or from negative to positive are accurately identified. Based on the location of the inflection points and the continuity of the power direction, the time period between adjacent inflection points is divided into the initial energy storage operation stage. The power direction is consistent in each stage, forming an initial segmentation of the energy storage charging and discharging behavior, which provides a basis for boundary correction and behavior attribute extraction.

[0025] By combining the rate of change of state of charge, power stability, and cumulative energy change, boundary corrections are performed on the energy storage operation phase to form an operating range that is time-continuous and physically consistent. For the initially defined energy storage operation phases, the rate of change of state of charge, power fluctuation amplitude, and cumulative energy change within each phase are calculated. The reasonableness of the phase boundaries is determined by setting reasonable thresholds or adaptive algorithms. For phases with discontinuous or abnormal boundaries, smooth boundary adjustment algorithms, dynamic interpolation, or methods to slightly extend / shorten the phase time can be used to ensure that each operating interval is continuous in time and physically conforms to the actual charging and discharging characteristics of energy storage, thus avoiding incorrect behavior division caused by noise or transient disturbances.

[0026] Extract the charging / discharging type identifier, duration parameter, energy change amplitude parameter, and power change characteristic parameter for each operating range; For each operating range after boundary correction, it is labeled as charging or discharging type according to the power direction. At the same time, the duration and cumulative energy change of the range are calculated. Parameters such as power average, power variance and maximum / minimum power change rate are obtained by integrating power and time. These parameters can not only quantify the charging and discharging behavior of energy storage, but also be used for behavior pattern analysis, anomaly detection and predictive modeling to ensure that each operating range is complete and traceable in terms of attributes.

[0027] Each operating interval is divided into a set of charging and discharging behavior segments with clear time boundaries and energy attribute identifiers; The extracted parameters and boundary information are integrated into a data structure of behavioral segments. Each segment contains identification information such as start time, end time, charging / discharging type, duration, energy change, and power characteristics. All operating intervals form a continuous set of segments in sequence, ensuring that the charging / discharging behavior of the system can be clearly represented in both time and energy dimensions, providing a set of directly usable behavioral data for monitoring, control, and optimization of energy storage systems.

[0028] S3: Based on the set of charging and discharging behavior segments, a multi-dimensional behavior association network structure of charging and discharging behavior segments is constructed. By analyzing the change characteristics of grid-connected power and load power during charging and discharging, an energy path mapping relationship between charging and discharging behavior segments and energy flow direction is established.

[0029] The process of constructing a multidimensional behavioral association network structure for charging and discharging behavior segments in S3 is as follows: The charging and discharging behavior segments are sequentially arranged according to time order to establish the temporal adjacency relationship between the behavior segments; The charging and discharging behavior segments are sorted from early to late according to their start time to form a continuous time series. For adjacent behavior segments in the series, their temporal adjacency is marked so that each segment can clearly correspond to the previous segment and the next segment. This forms node connections in the time dimension in network construction, providing a basic data structure for analyzing the continuity and causal relationship of charging and discharging behavior.

[0030] Based on the direction and amount of energy change of behavioral segments, establish the correlation of energy transfer of behavioral segments and construct the energy inheritance mapping relationship between behavioral segments; The charging or discharging type and energy change of each segment are identified. When the energy output or input of a segment may affect the energy state of the next segment, the energy change of the previous segment is transferred to the next segment in a direction through a mapping function. The transfer intensity can be represented by a weighted method. The weight can be adjusted based on the time interval, the proportion of energy change, or historical behavior patterns, thereby forming a quantifiable energy transfer network between segments.

[0031] By combining the grid-connected power change trend and load power response characteristics corresponding to the behavior segments, the correlation degree of energy regulation coordination between behavior segments is calculated. By analyzing the grid-connected power curves and load power responses during the occurrence of each segment, the power synchronization, amplitude matching degree, and trend similarity between segments are quantified. A multidimensional feature correlation matrix is ​​formed using Pearson correlation coefficient, dynamic time warping algorithm, or adaptive weight matching method. This matrix records multidimensional indicators of energy regulation consistency among segments, providing refined parameters for assigning multidimensional edge weights to the network.

[0032] Based on temporal adjacency, energy inheritance mapping, and regulatory coordination, a multidimensional behavior association network structure for charging and discharging behavior segments is constructed. Based on temporal adjacency, energy inheritance mapping, and synergy index in the multidimensional feature correlation matrix, a multidimensional behavior correlation network structure for charging and discharging behavior segments is constructed. Each charging and discharging behavior segment is treated as a network node, and multidimensional edges between nodes are established based on three types of relationships: temporal adjacency edges, energy inheritance edges, and synergistic adjustment edges. Each edge can be assigned a weight to represent the relationship strength. The weight of the synergistic adjustment edge is directly derived from the multidimensional feature correlation matrix. The network information is stored through a graph structure or adjacency matrix, enabling visualization of the multidimensional connections between various behavior segments in the energy storage system in terms of time, energy, and synergy. This provides a reliable data foundation and computational framework for operation mode analysis, abnormal behavior identification, and optimized control strategies.

[0033] The process of establishing the energy path mapping relationship between charging and discharging behavior segments and energy flow direction in S3 is as follows: In the multidimensional behavior association network structure, the combination relationship between each charging behavior segment and the corresponding discharging behavior segment is extracted; The nodes in the multidimensional behavioral association network are traversed to identify the behavioral fragment nodes of the charging type and the discharge type fragment nodes that may be affected by them. Based on the temporal adjacency relationship, energy inheritance edge and cooperative adjustment edge, the fragment pairs that can be combined are determined. Each combination relationship corresponds to the energy input and output association of the energy storage system in a specific time period, providing a basis for energy path analysis.

[0034] Based on the combination relationship of discharge behavior segments, the grid-connected power change, load power change and photovoltaic output change during each behavior segment are analyzed to identify the energy source node of energy storage and the target node of energy release. For each pair of charge-discharge segments, extract the power data of each node during its occurrence, including grid-connected power, load power, and photovoltaic output. By comparing the energy change and power change trend, determine the energy source node corresponding to the charging segment, such as photovoltaic output or grid-connected recharge, and the energy release target node of the discharging segment, such as load consumption or grid-connected transmission, thereby clarifying the starting and ending points of energy flow in the system.

[0035] Generate an energy flow sequence along the energy input / output path from the source node to the target node; Starting from the energy source node of the energy storage, the energy input and output of each node are recorded sequentially according to the combination relationship of charging-discharging segments and temporal adjacency relationship in the network, forming a continuous energy flow sequence. This not only includes the direction information of energy, but can also include the energy amplitude, power change characteristics and time stamp of each node, so as to realize a quantitative description of the entire process of energy flow in energy storage.

[0036] Each charging and discharging behavior segment is mapped to the corresponding energy flow sequence, forming an energy path mapping relationship between the charging and discharging behavior segment and the energy flow direction; The generated energy flow sequence is associated with the original charging and discharging behavior segment nodes to form a mapping table or mapping data structure between charging and discharging behavior segments and energy flow sequence. Each charging or discharging segment node corresponds to a unique energy flow sequence in the network. This sequence records the source, flow direction and amplitude change of energy during the segment. By establishing this one-to-one mapping relationship, the energy transfer path of each behavior segment in the system is clarified, and a direct traceable connection from behavior segment to energy flow is realized.

[0037] S4: Based on the energy path mapping relationship, an effective energy displacement criterion is introduced to evaluate the regulation contribution of associated charging and discharging behavior segments. When the charging and discharging behavior is highly coupled in the time dimension and no effective energy regulation displacement is formed, the energy storage operation is determined to be in a state of regulation contribution failure.

[0038] The process of evaluating the modulatory contribution of correlated charge-discharge behavior segments in S4 is as follows: Based on the energy path mapping relationship, the energy input path and energy output path corresponding to each charging and discharging behavior segment are extracted; The charging and discharging behavior of energy storage units is divided into several continuous or discrete behavior segments. Each segment corresponds to the input stage of energy storage charging and the output stage of energy discharging. Based on the energy path mapping relationship between the energy storage system, photovoltaic power generation system, load system and power grid, an energy path mapping model is constructed. Through topological graphs, matrices or graph theory, the energy input source, photovoltaic power generation, grid-connected input and output loads and grid-connected output of each segment are accurately identified, forming a set of traceable and quantifiable energy input and output paths, providing a clear data foundation for regulation contribution analysis.

[0039] For each path, energy time migration, energy spatial transfer, and energy regulation response contribution are constructed as evaluation factors for effective energy displacement criteria. After extracting the energy path mapping, a multidimensional energy migration analysis is performed on each path to calculate the energy time migration, i.e., the rate of change of energy storage charging and discharging at different time points. This is used to quantify the immediate effect of energy storage segments on smoothing the system power curve. Next, the energy spatial transfer is calculated, i.e., the distribution change of energy storage energy among different loads or grid-connected nodes. This is used to quantify the ability of energy storage to regulate the system power balance in the spatial dimension. Then, the energy regulation response contribution is evaluated by calculating the response efficiency of energy storage segments to load fluctuations, photovoltaic power generation fluctuations, and grid-connected power deviations, reflecting the effective energy displacement capability of energy storage segments. The above three factors serve as evaluation factors, providing a quantitative basis for the regulation contribution.

[0040] Based on the evaluation factors, the regulation effect of energy storage on system load changes, grid-connected power balance and photovoltaic fluctuation suppression in the energy input and output stages is analyzed, and the regulation contribution of each charging and discharging behavior segment to the system operation stability is calculated. Based on evaluation factors, the regulation effect of each charging and discharging segment is analyzed. For the energy input stage, the contribution of energy storage charging to load peak shaving, grid-connected power balance, and photovoltaic fluctuation smoothing is quantified. For the energy output stage, the regulation contribution of energy storage discharging to load support, low-valley load compensation, and photovoltaic power generation drop is quantified. System simulation or historical operation data playback can be used to compare the total load fluctuation rate, grid-connected power deviation, and photovoltaic power fluctuation amplitude before and after energy storage regulation. The regulation effect values ​​of each segment in the energy input and output stages are obtained. The regulation effect values ​​of each segment are combined with evaluation factors to establish a regulation contribution calculation model. Weighted integral or multi-index scoring methods can be used to assign weights to energy time migration, energy spatial transfer, and energy regulation response contribution to calculate the comprehensive regulation contribution of each charging and discharging segment. The weights can be flexibly adjusted according to the system operation objectives. For example, when strengthening photovoltaic fluctuation suppression, the weight of regulation response contribution is increased. The final value is the quantitative regulation contribution of the energy storage segment to the system operation stability, providing a reference for energy storage scheduling optimization.

[0041] Based on the magnitude of the regulatory contribution, the charging and discharging behavior segments are classified into value levels to form a regulatory contribution evaluation result; Based on the moderating contribution, each segment is divided into different value levels, such as high contribution, medium contribution, and low contribution, or classified according to a preset five-level scoring standard. Percentile division or threshold method can be used to accurately classify the contribution and generate complete evaluation results, including the moderating contribution value of each segment, the corresponding level, and energy path mapping information.

[0042] The process in S4 to determine that energy storage operation is in a state of regulatory contribution failure is as follows: Based on the evaluation results of the regulation contribution, charging and discharging behavior segments with regulation contribution below a preset threshold are selected. By analyzing the charging and discharging data of energy storage units over a certain period, the regulation contribution of each time segment is calculated. This contribution can be quantified by assessing the actual impact of energy storage behavior on load smoothing, peak-valley reduction, or mitigation of grid-connected power fluctuations. A pre-set threshold is used, determined based on historical operating data and system regulation targets; for example, it can be set to 30%–50% of the average regulation contribution. All charging and discharging segments below this threshold are marked as candidate segments, effectively filtering out energy storage behaviors that may not have played a regulatory role, providing fundamental data for subsequent diagnostics.

[0043] For the selected behavioral segments, analyze the correspondence between energy input and output of charging and discharging behaviors within a preset time window, and determine the effect of the correspondence on the improvement of the load curve. For each candidate segment, within a set time window, such as 10 minutes or 30 minutes, the energy input of the charging behavior and the energy output of the discharging behavior are matched according to the system load characteristics to form an input-output correspondence matrix. By analyzing the load curve or grid-connected power fluctuations, the actual improvement effect of the charging and discharging behavior on load peak reduction, valley filling or fluctuation smoothing is calculated. If the input-output correspondence fails to significantly improve the load curve, that is, the improvement effect is lower than the preset threshold, it can be combined with the standard deviation or volatility index, indicating that the regulation effect of the segment is not obvious, and it is necessary to proceed to the next step of energy space path analysis.

[0044] When the correspondence between energy input and output does not significantly improve the system's operating state, further analysis should be conducted on the spatial consistency between the energy input path and the energy output path. The energy flow of charging and discharging behavior on the physical or logical network is tracked to identify the location of the grid node, load node or energy storage unit corresponding to each charging and discharging, forming an input-output path mapping. By comparing the spatial distribution relationship between charging nodes and discharging nodes, the effectiveness of energy migration within the system is judged. If the energy input and output paths are highly consistent and point to the same or adjacent nodes, it indicates that the energy cycle is mainly limited to local nodes and has not effectively participated in the wide-area regulation of the system load, indicating a problem of insufficient spatial migration.

[0045] If the energy input path and the output path point to the same node, the energy spatial migration effect is deemed insufficient. Based on node mapping and spatial path analysis, the coverage area, migration distance, and degree of improvement on the system load curve of energy migration are calculated. If the energy input and output paths are concentrated in the same or adjacent nodes, and the improvement effect on load or grid-connected power fluctuation is not significant, it can be determined that the energy spatial migration effect of this segment is insufficient. This can be verified by combining historical operating data or simulation results to ensure that the judgment is accurate and reliable.

[0046] When energy storage behavior is characterized by low regulation contribution, energy circulation within the time window but without significantly improving load and grid-connected power fluctuations, and insufficient energy spatial migration effect, it is determined that the energy storage operation is in a state of regulation contribution failure. For each candidate segment, a comprehensive judgment is made based on the analysis results of the previous four steps. When the energy storage behavior simultaneously meets the following conditions: first, the regulation contribution is lower than the preset threshold; second, an energy cycle is formed within the preset time window, but the load and grid-connected power fluctuations are not significantly improved; and third, the energy input and output paths point to the same node, and the spatial migration effect is insufficient, the energy storage operation behavior is determined to be in a state of regulation contribution failure. The failure behavior can be marked and can be further used to optimize the energy storage scheduling strategy, such as adjusting the charging and discharging time, changing the energy flow direction, or redistributing energy storage resources, thereby improving the overall regulation efficiency of the system.

[0047] S5: Based on behavior triggering rules, the failure state of adjustment contribution is discretized into a set of structured operation behavior events, and the execution cycle of operation behavior events is matched and analyzed in combination with the characteristics of energy storage operation cycle to generate energy storage operation strategy adjustment prompt information.

[0048] In S5, the process of discretizing the failure state of adjustment contribution into a set of structured operational behavior events based on behavior triggering rules is as follows: Based on the combined characteristics of behavioral segments corresponding to the failure state of the adjustment contribution, the time start and end nodes, energy change amplitude and cycle number parameters of the failure behavior are extracted. The energy storage behavior segments determined to have failed in regulating contribution are analyzed. The start and end times of each segment are marked and recorded in the system time series. The energy change amplitude of the segment is further calculated, including the total charging energy, the total discharging energy, and the net energy difference, which serve as key parameters for measuring the energy cycle intensity. At the same time, the number of energy cycles, i.e., the number of charge and discharge cycles experienced by the stored energy within the time window, is counted to reflect the dynamic characteristics of energy storage behavior. The obtained time, energy, and cycle characteristics provide quantitative basis for clustering and event generation.

[0049] Clustering is performed on consecutive failure behaviors based on behavior triggering rules to generate runtime behavior events with unique identifiers; The extracted failure behavior feature vectors are used as input, and clustering analysis is performed in combination with preset behavior triggering rules, such as time continuity rules, energy threshold rules, and cycle number rules. Failure behavior segments with continuous or similar features are grouped into the same operational behavior event to reduce event redundancy and highlight key behavior patterns. Each generated operational behavior event is assigned a unique identifier, such as an event ID, to facilitate tracking, querying, and subsequent analysis. At the same time, the clustering algorithm can adopt methods such as hierarchical clustering based on distance metrics, density clustering, or time series similarity clustering to ensure the scientificity and consistency of event division.

[0050] Each operational behavior event is assigned a behavior type label, a failure level label, and an energy cycle characteristic label to form a structured set of operational behavior events; For each event, it is tagged based on its feature vector and clustering results. The behavior type tag is used to distinguish between charging-dominated, discharging-dominated, or cyclic failure behaviors. The failure level tag is comprehensively evaluated based on the decrease in regulation contribution, the number of energy cycles, and the load improvement effect. For example, it can be divided into mild failure, moderate failure, and severe failure. The energy cycle feature tag records the amplitude and frequency characteristics of energy cycles in the event. After the tag assignment is completed, all events are formed into a unified data structure, namely a structured set of operational behavior events, which facilitates statistical analysis, optimized scheduling, and risk assessment by the system. This set can be stored in the database in the form of tables or objects. Each event record contains a unique identifier, time range, energy parameters, number of cycles, and various tags, realizing complete information disclosure and facilitating computer processing.

[0051] The process of generating energy storage operation strategy adjustment prompts in S5 is as follows: Extract the operational characteristics of the energy storage system under daily, load, and photovoltaic power generation cycles; By collecting charging and discharging data, system load data, and photovoltaic power generation data of the energy storage system at different time periods, its operational characteristics can be analyzed. For the daily cycle, the charging and discharging patterns and peak-to-valley distribution of energy storage over 24 hours can be statistically analyzed, such as calculating the average hourly charging power, discharging power, and energy flow. For the load cycle, the frequency, amplitude, and time-series characteristics of energy storage charging and discharging behavior under typical load variation curves can be extracted, such as statistically analyzing the number of charging and discharging cycles and energy contribution when the load exceeds 80% of the average. For the photovoltaic power generation cycle, the energy response and regulation capabilities of energy storage under photovoltaic power variations can be analyzed, such as calculating the total energy discharged by energy storage during peak photovoltaic power periods and the peak-to-valley balance effect. Statistical analysis, time-series analysis, and frequency domain analysis methods, such as average curves, standard deviations, load response curves, and peak-to-valley ratio (peak power / valley power), can be used to form quantifiable periodic operational characteristics, providing a reference for identifying inefficient behavior.

[0052] By performing time-aligned analysis of the structured set of operational behavior events and the characteristics of each cycle, the recurrence pattern of inefficient behavior in different operational cycles can be identified. Align the cyclical characteristics with the structured set of operational behavior events generated in the previous stage according to the time series. Use a sliding time window method, for example, with a window length of 30 minutes and a window step size of 10 minutes, to map the occurrence time of each inefficient behavior event to the corresponding cyclical characteristic interval. By comparing the event time with the cyclical characteristics, statistically analyze the recurrence pattern of inefficient behavior in the daily cycle, load cycle, and photovoltaic power generation cycle. The matching rules can be set to a time tolerance of ±5 minutes and a power tolerance of ±5%. If the recurrence exceeds the threshold, such as more than 3 times in a day, it is marked as high-frequency inefficient behavior for risk assessment and strategy optimization.

[0053] The operational risk level is assessed based on the frequency of inefficient behavior, energy loss, and the degree of impact on system stability. Based on the patterns of repetitive behaviors, the frequency (number of times / day), cumulative energy loss, and contribution to improving load smoothness and grid-connected power fluctuations of each type of inefficient behavior are calculated. This can be measured by the percentage decrease in load fluctuation rate. Based on a comprehensive score, the operational risk level is divided into three levels: high, medium, and low. For example, a high-risk level is defined as a behavior frequency ≥ 3 times / day, energy loss ≥ 5 kWh, and load fluctuation improvement effect ≤ 10%; a medium-risk level is defined as a behavior frequency 1-2 times / day, energy loss 2-5 kWh, and load fluctuation improvement effect 10%-20%; and a low-risk level is defined as a behavior frequency < 1 time / day, energy loss < 2 kWh, and load fluctuation improvement effect > 20%. The score can be weighted using a weighting method, with weights of 0.4 for frequency, 0.3 for energy loss, and 0.3 for improvement effect, to obtain a comprehensive risk score, providing a quantitative basis for strategy adjustments.

[0054] Based on risk level analysis, the adjustment needs of the energy storage system are analyzed, and energy storage operation strategy adjustment prompts are generated. Based on the assessed risk level, the adjustment needs of the energy storage system in different cycles are analyzed to determine strategy directions such as optimizing charging and discharging times, adjusting energy distribution, or controlling the number of cycles. This generates adjustment prompts for energy storage operation strategies, including: suggested charging / discharging periods (e.g., high-load periods from 07:00 to 09:00), power adjustment range of 0.5 to 1MW, high, medium, and low priority of actions, and implementation precautions (e.g., avoiding more than two consecutive cycles). These prompts can be presented in structured data tables or JSON objects for direct access by system dispatchers or intelligent control modules. They can also be combined with historical operating data and simulation results to provide quantitative references, such as prompts like "Continuous inefficient behavior has increased the peak-valley difference by 5%, it is recommended to delay discharge by 1 hour," thereby improving energy storage operating efficiency and ensuring system stability.

[0055] Example 2: Please refer to Figure 2 As shown, a photovoltaic energy storage charging and discharging behavior monitoring system includes: Energy modeling module: Synchronously collects data from energy storage, photovoltaics, grid connection and load, and constructs a continuous time-series energy operation model of the overall energy flow; Behavior decomposition module: Based on the power change and state of charge evolution laws, the energy storage operation process is decomposed into a set of charging and discharging behavior segments with time boundaries and energy attribute labels; Network association module: Constructs a multi-dimensional behavior association network based on behavior segments, and forms an energy path mapping relationship between charging and discharging behavior segments and energy flow direction; Contribution evaluation module: Utilizes energy path mapping relationship and effective displacement criterion to analyze the moderating contribution of behavioral segments and identify the failure state of moderating contribution; Strategy prompt module: Discretizes the adjustment failure state into structured operation events, and generates energy storage operation strategy adjustment prompt information in combination with the operation cycle.

[0056] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0057] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring the charging and discharging behavior of photovoltaic energy storage, characterized in that, Includes the following steps: By synchronously collecting energy operation data from energy storage units, photovoltaic power generation side, grid connection side and load side, and performing reconstruction processing on multi-sampling period data based on a unified time reference, a continuous time-series energy operation model of the overall energy flow state is constructed. In the continuous time-series energy operation model, based on the characteristics of energy storage power direction change and the dynamic evolution law of state of charge, the execution behavior of the energy storage operation process is structurally decomposed to form a set of charging and discharging behavior segments with time boundaries and energy attribute identifiers. Based on the set of charging and discharging behavior segments, a multi-dimensional behavior association network structure of charging and discharging behavior segments is constructed. By analyzing the change characteristics of grid-connected power and load power during charging and discharging, an energy path mapping relationship between charging and discharging behavior segments and energy flow direction is established. Based on the energy path mapping relationship, an effective energy displacement criterion is introduced to evaluate the regulation contribution of associated charging and discharging behavior segments. When the charging and discharging behavior is highly coupled in the time dimension and no effective energy regulation displacement is formed, the energy storage operation is determined to be in a state of regulation contribution failure. Based on behavior triggering rules, the failure state of adjustment contribution is discretized into a set of structured operation behavior events. Combined with the characteristics of energy storage operation cycle, the execution cycle of operation behavior events is matched and analyzed to generate energy storage operation strategy adjustment prompt information.

2. The method for monitoring the charging and discharging behavior of photovoltaic energy storage according to claim 1, characterized in that, The process of constructing a continuous time-series energy operation model of the overall energy flow state is as follows: The charging and discharging power data, state of charge data, and battery operating status data of the energy storage unit are collected in real time. At the same time, the output power data of the photovoltaic power generation side, the switching power data of the grid-connected side, and the energy consumption power data of the load side are acquired synchronously. Time alignment is performed on data with different sampling frequencies based on a unified time reference; By constructing a time index mapping relationship, missing sampled data is interpolated and repaired, and abnormal fluctuation data is denoised and smoothed to form a data sequence with a unified time granularity. Based on the energy conservation principle, the energy exchange process of the energy storage side, power generation side, grid connection side and load side is checked for consistency of flow direction, and a multi-node energy flow coupling relationship matrix is ​​constructed. Based on the coupling relationship matrix, time-series fusion processing is performed on the energy exchange data of multiple nodes to generate a continuous time-series energy operation model that reflects the overall energy input, storage, output and consumption relationship of the system.

3. The method for monitoring the charging and discharging behavior of photovoltaic energy storage according to claim 2, characterized in that, The process of forming a set of charging and discharging behavior segments with temporal boundaries and energy attribute identifiers is as follows: In the continuous time-series energy operation model, identify the inflection points of positive and negative changes in energy storage power, and divide the energy storage operation stage according to the continuous interval of power direction; By combining the rate of change of state of charge, power stability, and cumulative energy change, boundary corrections are performed on the energy storage operation phase to form an operating range that is time-continuous and physically consistent. Extract the charging / discharging type identifier, duration parameter, energy change amplitude parameter, and power change characteristic parameter for each operating range; Each operating interval is divided into a set of charging and discharging behavior segments with clear time boundaries and energy attribute identifiers.

4. The method for monitoring the charging and discharging behavior of photovoltaic energy storage according to claim 3, characterized in that, The process of constructing a multidimensional behavioral association network structure for charging and discharging behavior segments is as follows: The charging and discharging behavior segments are sequentially arranged according to time order to establish the temporal adjacency relationship between the behavior segments; Based on the direction and amount of energy change of behavioral segments, establish the correlation of energy transfer of behavioral segments and construct the energy inheritance mapping relationship between behavioral segments; By combining the grid-connected power change trend and load power response characteristics corresponding to the behavior segments, the correlation degree of energy regulation coordination between behavior segments is calculated. Based on temporal adjacency, energy inheritance mapping, and regulatory coordination, a multidimensional behavioral association network structure for charging and discharging behavior segments is constructed.

5. The method for monitoring the charging and discharging behavior of photovoltaic energy storage according to claim 4, characterized in that, The process of establishing the energy path mapping relationship between charging / discharging behavior segments and energy flow direction is as follows: In the multidimensional behavior association network structure, the combination relationship between each charging behavior segment and the corresponding discharging behavior segment is extracted; Based on the combination relationship of discharge behavior segments, the grid-connected power change, load power change and photovoltaic output change during each behavior segment are analyzed to identify the energy source node of energy storage and the target node of energy release. Generate an energy flow sequence along the energy input / output path from the source node to the target node; Each charging and discharging behavior segment is mapped to its corresponding energy flow sequence, forming an energy path mapping relationship between the charging and discharging behavior segments and the energy flow direction.

6. The method for monitoring the charging and discharging behavior of photovoltaic energy storage according to claim 5, characterized in that, The process of evaluating the moderating contribution of correlated charge and discharge behavior segments is as follows: Based on the energy path mapping relationship, the energy input path and energy output path corresponding to each charging and discharging behavior segment are extracted; For each path, energy time migration, energy spatial transfer, and energy regulation response contribution are constructed as evaluation factors for effective energy displacement criteria. Based on the evaluation factors, the regulation effect of energy storage on system load changes, grid-connected power balance and photovoltaic fluctuation suppression in the energy input and output stages is analyzed, and the regulation contribution of each charging and discharging behavior segment to the system operation stability is calculated. Based on the magnitude of the regulatory contribution, the charging and discharging behavior segments are classified into value levels to form a regulatory contribution evaluation result.

7. The method for monitoring the charging and discharging behavior of photovoltaic energy storage according to claim 6, characterized in that, The process of determining that energy storage operation is in a state of regulatory contribution failure is as follows: Based on the evaluation results of the regulation contribution, charging and discharging behavior segments with regulation contribution below a preset threshold are selected. For the selected behavioral segments, analyze the correspondence between energy input and output of charging and discharging behaviors within a preset time window, and determine the effect of the correspondence on the improvement of the load curve. When the correspondence between energy input and output does not significantly improve the system's operating state, further analysis should be conducted on the spatial consistency between the energy input path and the energy output path. If the energy input path and the output path point to the same node, the energy spatial migration effect is deemed insufficient. When energy storage behavior exhibits both low regulation contribution and energy circulation within the time window without significantly improving load and grid-connected power fluctuations, and the energy spatial migration effect is insufficient, it is determined that the energy storage operation is in a state of regulation contribution failure.

8. The method for monitoring the charging and discharging behavior of photovoltaic energy storage according to claim 7, characterized in that, The process of discretizing the failure state of adjustment contribution into a set of structured operational behavior events based on behavior triggering rules is as follows: Based on the combined characteristics of behavioral segments corresponding to the failure state of the adjustment contribution, the time start and end nodes, energy change amplitude and cycle number parameters of the failure behavior are extracted. Clustering is performed on consecutive failure behaviors based on behavior triggering rules to generate runtime behavior events with unique identifiers; Each operational behavior event is assigned a behavior type label, a failure level label, and an energy cycle characteristic label to form a structured set of operational behavior events.

9. The method for monitoring the charging and discharging behavior of photovoltaic energy storage according to claim 8, characterized in that, The process of generating energy storage operation strategy adjustment prompts is as follows: Extract the operational characteristics of the energy storage system under daily, load, and photovoltaic power generation cycles; By performing time-aligned analysis of the structured set of operational behavior events and the characteristics of each cycle, the recurrence pattern of inefficient behavior in different operational cycles can be identified. The operational risk level is assessed based on the frequency of inefficient behavior, energy loss, and the degree of impact on system stability. Based on risk level analysis, the adjustment needs of the energy storage system are analyzed, and information prompts for adjusting the energy storage operation strategy are generated.

10. A photovoltaic energy storage charging and discharging behavior monitoring system, applied to the method described in any one of claims 1-9, characterized in that, include: Energy modeling module: Synchronously collects data from energy storage, photovoltaics, grid connection and load, and constructs a continuous time-series energy operation model of the overall energy flow; Behavior decomposition module: Based on the power change and state of charge evolution laws, the energy storage operation process is decomposed into a set of charging and discharging behavior segments with time boundaries and energy attribute labels; Network association module: Constructs a multi-dimensional behavior association network based on behavior segments, and forms an energy path mapping relationship between charging and discharging behavior segments and energy flow direction; Contribution evaluation module: Utilizes energy path mapping relationship and effective displacement criterion to analyze the moderating contribution of behavioral segments and identify the failure state of moderating contribution; Strategy prompt module: Discretizes the adjustment failure state into structured operation events, and generates energy storage operation strategy adjustment prompt information in combination with the operation cycle.

Images