Energy scheduling method and platform system of photovoltaic energy storage system

Abstract

The application relates to an energy scheduling method and platform system of a photovoltaic energy storage system, and the method comprises the following steps: acquiring working state data of the photovoltaic energy storage system at multiple historical time points through a sensor; training a state prediction model based on the working state data corresponding to each historical time point; predicting a predicted scheduling demand at a current time point based on the state prediction model and the working state data at a previous time point; and determining an energy scheduling instruction corresponding to the photovoltaic energy storage system according to the predicted scheduling demand in response to an energy scheduling demand signal at the current time point.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy storage system control, in particular to an energy scheduling method and platform system of a photovoltaic energy storage system. BACKGROUND

[0002] With the rapid development of renewable energy technology, photovoltaic power generation as a new type of energy has been widely used in the world. However, photovoltaic energy storage systems still face many challenges in actual operation, especially under complex environmental and variable load conditions. How to achieve efficient energy scheduling and optimal management remains a difficult point in technology development.

[0003] The existing solutions lack time series analysis of historical load data and dynamic prediction of equipment working characteristics, making it difficult to accurately identify scheduling requirements. Common rule mechanisms or empirical formulas cannot adapt to diversified energy scheduling scenarios, resulting in insufficient precision of scheduling strategy generation, which easily causes equipment delay response, limiting the intelligent level and operation safety of photovoltaic energy storage systems. Therefore, the existing technology has defects and needs to be solved. SUMMARY

[0004] The primary purpose of the present application is to solve at least one of the above problems and provide an energy scheduling method and platform system of a photovoltaic energy storage system.

[0005] To achieve the various purposes of the present application, the present application adopts the following technical solutions:

[0006] An energy scheduling method of a photovoltaic energy storage system is provided to adapt to one of the purposes of the present application, comprising the following steps:

[0007] Obtain working state data of the photovoltaic energy storage system at a plurality of historical time points through a sensor;

[0008] Train a state prediction model based on the working state data corresponding to each of the historical time points;

[0009] Based on the state prediction model, predict the predicted scheduling requirement of the current time point according to the working state data of the previous time point;

[0010] In response to the energy scheduling requirement signal of the current time point, determine the energy scheduling instruction corresponding to the photovoltaic energy storage system according to the predicted scheduling requirement.

[0011] In an optional embodiment, training a state prediction model based on the working state data corresponding to each of the historical time points comprises:

[0012] Determine the state characteristic parameters corresponding to each of the working state data based on a state identification rule;

[0013] sequencing each of the state characteristic parameters from early to late based on the historical time points, to obtain a parameter time sequence;

[0014] training a preset long short-term memory neural network using the parameter time sequence as a training data set, to obtain a state prediction model.

[0015] In an optional embodiment, the working state data includes illumination data, photovoltaic power generation power, energy storage battery voltage, energy storage battery current, energy storage battery temperature, energy storage battery state, and electricity price data; the state characteristic parameters include illumination intensity, photovoltaic power generation capacity, energy storage capacity, load demand, real-time electricity price, load type, and electricity price type; the power grid load type includes power generation capacity greater than the sum of energy storage capacity demand and user load demand, power generation capacity less than user load demand, and power generation capacity greater than or equal to user load demand; and the electricity price type includes off-peak electricity price and peak electricity price.

[0016] In an optional embodiment, training a preset long short-term memory neural network using the parameter time sequence as a training data set, to obtain a state prediction model, includes:

[0017] calculating a difference value between the illumination intensities of any two adjacent state characteristic parameters in the parameter time sequence, to obtain an illumination difference value;

[0018] calculating a difference value between the photovoltaic power generation capacities of the two adjacent state characteristic parameters, to obtain a power generation difference value;

[0019] calculating a difference value between the energy storage capacities of the two adjacent state characteristic parameters, to obtain an energy storage difference value;

[0020] calculating a difference value between the load demands of the two adjacent state characteristic parameters, to obtain a load difference value;

[0021] calculating a difference value between the real-time electricity prices of the two adjacent state characteristic parameters, to obtain an electricity price difference value;

[0022] calculating a weighted sum of the illumination difference value, the power generation difference value, the energy storage difference value, the load difference value, and the electricity price difference value, to obtain a change characteristic parameter;

[0023] calculating a ratio of the change characteristic parameter to a time difference value between the historical time points corresponding to the two state characteristic parameters, to obtain a correlation characteristic between the two state characteristic parameters;

[0024] concatenating the parameter time sequence and each corresponding correlation characteristic as a fusion characteristic, and inputting the fusion characteristic to a preset long short-term memory neural network for training, to obtain a state prediction model.

[0025] In an optional embodiment, the loss function of the long short-term memory neural network is set as the product of a first cross-entropy, a second cross-entropy, and a third cross-entropy. The first cross-entropy is the difference between the predicted output and the label corresponding to the parameter time series; the second cross-entropy is the difference between the output association feature and the actual association feature; the third cross-entropy is the difference between the predicted output and the scheduling demand representation of the output association feature and the real label corresponding to the actual scheduling demand; the output association feature is the association feature between the predicted output and the predicted output of the previous time step; and the actual association feature is the actual association feature between the corresponding parameters of the parameter time series.

[0026] In an optional embodiment, the sensor includes a light sensor, a temperature sensor, a battery status sensor, and a system status sensor. The light sensor is used to detect ambient light intensity, the temperature sensor is used to detect the operating temperature of the photovoltaic energy storage system, the battery status sensor is used to detect the voltage, current, and state of charge of the energy storage battery in the energy storage system, and the system status sensor is used to identify the load type and electricity price type based on the operating status data.

[0027] In an optional embodiment, based on the state prediction model and the work state data from the previous time point, the predicted scheduling requirements for the current time point are predicted, including:

[0028] Obtain the work status data from at least the two previous time points of the current time point and identify them as historical status data;

[0029] Calculate the correlation features between the two historical state data;

[0030] The historical state data and corresponding associated features are input into the state prediction model to obtain the predicted scheduling requirements at the current time point.

[0031] In an optional embodiment, determining the energy dispatch command corresponding to the photovoltaic energy storage system based on the predicted dispatch demand includes:

[0032] Obtain the current working status data and corresponding scheduling requirement parameters corresponding to the energy scheduling demand signal;

[0033] Determine the reference state features corresponding to the predicted scheduling demand in the preset database corresponding to the scheduling demand parameters;

[0034] Calculate the feature similarity between the current working state data and the reference state features;

[0035] Determine whether the feature similarity is greater than a preset similarity threshold;

[0036] If not, the energy dispatch command corresponding to the photovoltaic energy storage system is determined through the actual user interface operation of the dispatch platform;

[0037] If so, based on the predicted scheduling requirements and the correspondence between preset working states and scheduling strategies, the energy scheduling instructions for the photovoltaic energy storage system are determined.

[0038] In an optional embodiment, the scheduling strategy includes:

[0039] When there is sufficient sunlight, after the load power supply is balanced, the photovoltaic system's power generation will first supply the energy storage battery for charging, and then supply the grid.

[0040] When sunlight is insufficient, the photovoltaic system prioritizes power generation to supply user loads, and the energy storage system prioritizes power discharge to supply user loads. If power generation and discharge cannot meet the user load demand, then the grid is the next priority to purchase electricity.

[0041] When the power grid fails, the photovoltaic energy storage system switches to off-grid mode, prioritizing power supply to user loads through photovoltaic power generation and energy storage discharge;

[0042] During off-peak electricity prices, electricity is purchased from the grid at a low price to charge the energy storage system;

[0043] During peak electricity prices, the energy storage system supplies power to user loads and discharges to the grid for sale at higher prices.

[0044] On the other hand, an energy dispatch platform system for a photovoltaic energy storage system provided to meet one of the purposes of this application includes:

[0045] The acquisition module is used to acquire operational status data of the photovoltaic energy storage system at multiple historical time points through sensors;

[0046] The training module is used to train a state prediction model based on the working state data corresponding to each of the historical time points.

[0047] The prediction module is used to predict the scheduling requirements at the current time point based on the state prediction model and the work status data at the previous time point.

[0048] The determination module is used to respond to the energy dispatch demand signal at the current time point and determine the energy dispatch command corresponding to the photovoltaic energy storage system based on the predicted dispatch demand.

[0049] On another front, an energy dispatching device for a photovoltaic energy storage system, provided to meet one of the purposes of this application, includes a central processing unit and a memory, wherein the central processing unit is used to invoke and run a computer program stored in the memory to execute the steps of the energy dispatching method for the photovoltaic energy storage system described in this application.

[0050] In another aspect, a computer-readable storage medium is provided to suit one of the purposes of this application, the computer-readable storage medium storing computer-executable instructions for causing a computer to perform an energy dispatching method for a photovoltaic energy storage system as disclosed in any of the first aspects of this invention.

[0051] The technical solution of this application has many advantages, including but not limited to the following aspects:

[0052] This application uses sensors to acquire historical time-point data of the photovoltaic energy storage system to train a state prediction model. Based on the data from the previous time point, it predicts the scheduling needs at the current time point and further determines the energy scheduling instructions. This enables the generation of accurate energy scheduling instructions based on the time-series analysis of the working state, thereby improving the accuracy and operational safety of intelligent energy scheduling control of the photovoltaic energy storage system. Attached Figure Description

[0053] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0054] Figure 1 This is a schematic diagram of the architecture between the photovoltaic energy storage system and the energy dispatch platform system of this application;

[0055] Figure 2 for Figure 1 A schematic diagram of another architecture;

[0056] Figure 3 A flowchart illustrating one embodiment of the energy dispatching method for the photovoltaic energy storage system of this application;

[0057] Figure 4 This is a schematic diagram of the energy dispatch platform system of the photovoltaic energy storage system used in this application. Detailed Implementation

[0058] The technical solution of this application is applicable to the field of energy storage system control technology, and is particularly applicable to the energy dispatching scenario of photovoltaic energy storage system. In this context, the technical solution of this application can be applied in a typical energy dispatching platform system architecture.

[0059] like Figure 1The embodiment of this disclosure provides an energy dispatching platform system 40 for a photovoltaic energy storage system connected to a photovoltaic energy storage system 10. The photovoltaic energy storage system 10 includes a photovoltaic system 20 for power generation and an energy storage system 30 for storing electricity. The photovoltaic system 20 includes several photovoltaic arrays, and the energy storage system 30 includes several energy storage batteries. The energy dispatching platform system 40 receives the working status data uploaded by the photovoltaic energy storage system and issues a control for real-time control of the energy flow between the energy storage system, the photovoltaic system, the power grid, and user loads, thereby ensuring the efficient operation of the photovoltaic energy storage system.

[0060] The energy dispatching platform system 40 of the photovoltaic energy storage system is used to perform the following steps: acquiring the working status data of the photovoltaic energy storage system at multiple historical time points through sensors; training a state prediction model based on the working status data corresponding to each historical time point; predicting the predicted dispatching demand at the current time point based on the state prediction model and the working status data at the previous time point; and determining the energy dispatching instruction corresponding to the photovoltaic energy storage system according to the predicted dispatching demand in response to the energy dispatching demand signal at the current time point.

[0061] It should be noted that the energy dispatching platform system 40 in this embodiment can be a local device, a remote device, or a combination of both.

[0062] Please see Figure 2 This disclosure discloses another architecture between a photovoltaic energy storage system and an energy dispatch platform system. The energy dispatch platform system 40 of the photovoltaic energy storage system includes: an edge computing platform 50, used to acquire the working status data of the photovoltaic energy storage system, predict the energy dispatch demand corresponding to the working status at a target time point using a state prediction model, and determine the energy dispatch instructions for the photovoltaic energy storage system based on the predicted dispatch demand; an energy dispatch database 60, used to store working status data, state characteristic parameters, state identification rules, correlation features of historical state data, dispatch demand parameters, reference state characteristics, preset thresholds, state prediction models, and training data corresponding to the state prediction models, and to provide the energy dispatch data and models to the edge computing platform 50; and a cloud management platform 70, used to visualize the data in the edge computing platform and the energy dispatch database 60, issue control instructions to all devices in the photovoltaic energy storage system, and manually control the energy dispatch of the photovoltaic energy storage system based on human experience and prior knowledge.

[0063] The energy dispatching method of a photovoltaic energy storage system disclosed in this application can be programmed into a computer program product and deployed on a server for execution. For example, in an exemplary application scenario of this application, it can be deployed on the server of an energy dispatching platform. In this way, the method can be executed by human-computer interaction with the process of the computer program product through a graphical user interface by accessing the interface opened after the computer program product is run.

[0064] The specific embodiments described below can be combined with each other, and the same or similar concepts or processes will not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0065] Please see Figure 3 This application discloses an energy dispatching method for a photovoltaic energy storage system, comprising the following steps:

[0066] Step 1100: Acquire the operating status data of the photovoltaic energy storage system at multiple historical time points through sensors;

[0067] Optionally, the sensor may be a temperature sensor, a current sensor, a voltage sensor, a power sensor, a humidity sensor, or a combination of multiple sensors; this application does not impose any limitations.

[0068] Optionally, the acquisition of working status data can be achieved through real-time acquisition, periodic sampling, triggered acquisition, or low-power acquisition, and this application does not impose any limitations on this.

[0069] Step 2100: Based on the working status data corresponding to each of the historical time points, train a status prediction model;

[0070] Optionally, the state prediction model can be a long short-term memory network model, a Transformer model, a recurrent neural network model, or a combination of at least two models; this application does not impose any limitations.

[0071] Optionally, the training process of the state prediction model can be based on self-supervised learning, semi-supervised learning, or reinforcement learning, and this application does not impose any restrictions.

[0072] Optionally, the state prediction model can be optimized by combining sensor characteristics, equipment characteristics of the photovoltaic energy storage system, time-series electricity price variation characteristics, and time-series load variation characteristics; this application does not impose any limitations on this.

[0073] Step 3100: Based on the state prediction model, predict the scheduling requirements at the current time point according to the work state data at the previous time point;

[0074] Optionally, the predicted dispatch demand may include the energy dispatch demand during grid outages, the energy dispatch demand during periods of insufficient sunlight, the energy dispatch demand during periods of sufficient sunlight, the energy storage charging demand during periods of low load, the energy storage discharging demand during periods of high load, the electricity purchase demand during periods of low off-peak electricity prices, and the electricity sales demand during periods of peak electricity prices. This application does not impose any limitations on these requirements.

[0075] Optionally, the prediction process of the state prediction model can be implemented based on real-time inference, batch working state data processing, or incremental calculation, and this invention does not limit it.

[0076] Step 4100: In response to the energy dispatch demand signal at the current time point, determine the energy dispatch command corresponding to the photovoltaic energy storage system based on the predicted dispatch demand.

[0077] Optionally, the energy dispatch demand signal can be a sensor trigger signal, a local management interaction signal, a cloud management interaction signal, or other external input signal, and this application does not limit it.

[0078] Optionally, energy dispatch instructions may include energy storage charging, energy storage discharging, grid power purchase, grid power sale, priority supply to user loads, priority energy storage, and priority power sale, etc., which are not limited in this application.

[0079] As can be seen, the above embodiments train a state prediction model by acquiring the working state data of the photovoltaic energy storage system at different historical times through sensors, predict the current scheduling demand based on the previous time data, and further determine the energy scheduling instructions, thereby realizing the generation of accurate energy scheduling instructions based on the time-series analysis of working state data, and improving the accuracy and operational safety of the intelligent energy scheduling operation of the photovoltaic energy storage system.

[0080] In practical implementation, the step above, training a state prediction model based on the work status data corresponding to each historical time point, includes:

[0081] Based on the state recognition rules, determine the state feature parameters corresponding to each of the working state data;

[0082] Based on the historical time points, each of the state feature parameters is sorted from morning to evening to obtain a parameter time series;

[0083] Using the time series of the parameters as a training dataset, a preset long short-term memory neural network is trained to obtain a state prediction model.

[0084] Optionally, the long short-term memory neural network can be a single-layer network, a multi-layer network, a bidirectional network, or a network with an embedded attention mechanism; this application does not impose any limitations on it.

[0085] As can be seen, through the above specific embodiments, by extracting state feature parameters based on state recognition rules and generating parameter time series by sorting them by time, a long short-term memory network is trained to obtain a state prediction model. Thus, based on the generation of accurate energy regulation instructions, the accuracy of state prediction is improved through time series feature extraction and model training, providing reliable scheduling demand support for energy scheduling instructions, reducing the risk of misjudgment of energy scheduling instructions due to deviations caused by state predictions deviating from actual scheduling needs, and improving the operational safety of photovoltaic energy storage system energy scheduling.

[0086] In a specific embodiment, the working status data in the above steps includes sunlight data, photovoltaic power generation, energy storage battery voltage, energy storage battery current, energy storage battery temperature, energy storage battery status, and electricity price data; the status characteristic parameters include sunlight intensity, photovoltaic power generation, energy storage capacity, load demand, real-time electricity price, load type, and electricity price type; the grid load type includes power generation greater than the sum of energy storage capacity demand and user load demand, power generation less than user load demand, and power generation greater than or equal to user load demand; the electricity price type includes off-peak electricity price and peak electricity price.

[0087] As can be seen, the specific embodiments described above define the content of the state characteristic parameters to comprehensively characterize the working state characteristics of the energy flow process between the energy storage system, the photovoltaic system, the power grid, and the user load. This assists in generating accurate energy dispatching instructions based on state timing analysis, thereby improving the accuracy and operational safety of intelligent energy dispatching operations of the photovoltaic energy storage system.

[0088] In practical implementation, the above steps involve using the parameter time series as a training dataset to train a preset long short-term memory neural network to obtain a state prediction model, including:

[0089] For any two adjacent state feature parameters within the parameter time series, calculate the difference between the light intensity of the two adjacent state feature parameters to obtain the light difference.

[0090] Calculate the difference in photovoltaic power generation between two adjacent state characteristic parameters to obtain the power generation difference;

[0091] Calculate the difference in energy storage capacity between two adjacent state characteristic parameters to obtain the energy storage difference;

[0092] Calculate the difference between the load requirements of two adjacent state characteristic parameters to obtain the load difference;

[0093] Calculate the difference between the real-time electricity prices of two adjacent state characteristic parameters to obtain the electricity price difference;

[0094] The weighted sum of the difference in illumination, the difference in power generation, the difference in energy storage, the difference in load, and the difference in electricity price is calculated to obtain the characteristic parameters of change.

[0095] Calculate the ratio of the time difference between the change feature parameter and the historical time points corresponding to the two state feature parameters to obtain the correlation feature between the two state feature parameters;

[0096] The time series of the parameters and each of the corresponding associated features are concatenated into a fusion feature, which is then input into a preset long short-term memory neural network for training to obtain a state prediction model.

[0097] Optionally, the weighted summation can be calculated using fixed weights, dynamic weights, or adaptive weights, and this application does not impose any restrictions.

[0098] Optionally, the fused features can be multi-dimensional feature vectors, sequence feature matrices, or mixed feature sets, which are not limited in this application.

[0099] Optionally, the process of splicing the parameter time series and associated features can be achieved based on feature splicing, data alignment or dimensional transformation, and this application does not impose any limitations.

[0100] Optionally, the training process of the Long Short-Term Memory Network can be optimized by combining data augmentation, sequence length optimization, or model regularization, which is not limited in this application.

[0101] As can be seen, through the above specific embodiments, the changing characteristic parameters are obtained by weighted summation of the differences in power generation, energy storage, load, and electricity price among adjacent state characteristic parameters, and then combined with the time difference to generate associated features and splice them into fused features to train the long short-term memory network. Thus, on the basis of training an accurate state prediction model, the model's ability to capture state changes is improved through multi-dimensional feature analysis and associated feature fusion, providing more accurate prediction support for the generation of energy dispatch instructions and reducing the risk of state prediction error.

[0102] In specific implementation, in the above steps, the loss function of the long short-term memory neural network is set as the product of the first cross-entropy, the second cross-entropy, and the third cross-entropy. The first cross-entropy is the difference between the predicted output and the label corresponding to the parameter time series; the second cross-entropy is the difference between the output association feature and the actual association feature; the third cross-entropy is the difference between the predicted output and the scheduling demand representation of the output association feature and the real label corresponding to the actual scheduling demand; the output association feature is the association feature between the predicted output and the predicted output of the previous time step; and the actual association feature is the actual association feature between the corresponding parameters of the parameter time series.

[0103] Optionally, the loss function of the long short-term memory neural network can be implemented by combining regularization terms, weight adjustments, or dynamic loss balancing, and this application does not impose any restrictions.

[0104] Optionally, the first cross-entropy, the second cross-entropy, and the third cross-entropy can be categorical cross-entropy, regression cross-entropy, or mixed cross-entropy; this application does not impose any limitations on these.

[0105] As can be seen, through the above specific embodiments, by designing the loss function of the Long Short-Term Memory network as the product of the first cross-entropy, the second cross-entropy, and the third cross-entropy, the model's prediction accuracy for state sequences and associated features is improved through triple cross-entropy optimization, thereby providing more reliable model support for the generation of energy scheduling instructions and reducing the risk of prediction bias caused by insufficient model training.

[0106] In specific implementation, the sensors in the above steps include a light sensor, a temperature sensor, a battery status sensor, and a system status sensor. The light sensor is used to detect the ambient light intensity, the temperature sensor is used to detect the operating temperature of the photovoltaic energy storage system, the battery status sensor is used to detect the voltage, current, and state of charge of the energy storage battery in the energy storage system, and the system status sensor is used to identify the load type and electricity price type based on the operating status data.

[0107] Optionally, the battery status sensor and system status sensor can be a combination of sensors and control elements, wherein the control element can be an embedded microcontroller, a single-chip microcomputer or other dedicated chip, and the control element has a built-in detection algorithm to identify the battery status and system status at the current time point based on the operating status data.

[0108] As can be seen, through the above specific implementation methods, the comprehensiveness and accuracy of state feature extraction are improved by multi-sensor collaboration and system state analysis, providing high-quality data input for the state prediction model and reducing the risk of recognition errors caused by insufficient data from a single sensor.

[0109] In practical implementation, the above steps, based on the state prediction model and the work state data from the previous time point, predict the scheduling requirements for the current time point, including:

[0110] Obtain the work status data from at least the two previous time points of the current time point and identify them as historical status data;

[0111] Calculate the correlation features between the two historical state data;

[0112] The historical state data and corresponding associated features are input into the state prediction model to obtain the predicted scheduling requirements at the current time point.

[0113] As can be seen, through the above optional embodiments, by acquiring historical state data from at least two time points prior to the current time point and calculating correlation features, the data is input into the state prediction model to predict the operating state characteristics of the equipment in the current photovoltaic energy storage system. Thus, based on the generation of precise energy dispatch instructions, the timing accuracy of the dispatch demand predicted based on the state characteristics is improved through historical data and correlation feature analysis, providing a precise basis for energy dispatch instructions, reducing the risk of prediction errors caused by insufficient timing data, and improving the operational safety of energy dispatch in the photovoltaic energy storage system.

[0114] In practical implementation, the above steps, including determining the energy dispatch command corresponding to the photovoltaic energy storage system based on the predicted dispatch demand, include:

[0115] Obtain the current working status data and corresponding scheduling requirement parameters corresponding to the energy scheduling demand signal;

[0116] Determine the reference state features corresponding to the predicted scheduling demand in the preset database corresponding to the scheduling demand parameters;

[0117] Calculate the feature similarity between the current working state data and the reference state features;

[0118] Determine whether the feature similarity is greater than a preset similarity threshold;

[0119] If not, the energy dispatch command corresponding to the photovoltaic energy storage system is determined through the actual user interface operation of the dispatch platform;

[0120] If so, based on the predicted scheduling requirements and the correspondence between preset working states and scheduling strategies, the energy scheduling instructions for the photovoltaic energy storage system are determined.

[0121] Optionally, the feature similarity can be cosine similarity, Euclidean distance, Jaccard system or dynamic time warping distance, which is not limited in this application.

[0122] Optionally, the calculation of feature similarity can be based on vector comparison, statistical analysis, or feature matching, and this application does not impose any restrictions.

[0123] Optionally, the current operating status data may include power generation data, energy storage data, load data, or electricity price data, which are not limited in this application.

[0124] Optionally, the dispatch demand parameters may include generation parameters, discharge parameters, energy storage parameters, power purchase parameters, power sales parameters, or energy dispatch priority parameters, which are not limited in this application.

[0125] Optionally, the process of acquiring the current working status data and the corresponding scheduling requirement parameters can be based on direct sensor acquisition, controller processing, database calls, or data interface transmission, and this application does not impose any limitations.

[0126] Optionally, the actual user interface operation through the scheduling platform can be a local input operation or a cloud input operation, and this application does not limit it.

[0127] Optionally, the correspondence between working status and scheduling strategy can be represented by a mapping table, rule base, data-driven model or conditional logic table, which is not limited in this application.

[0128] As can be seen, through the above specific embodiments, by comparing the similarity between the state characteristics of the current working state data and the reference state characteristics, and determining the energy dispatch instructions based on the correspondence between predicted dispatch requirements and dispatch strategies, the accuracy and reliability of energy dispatch instruction generation are improved through feature similarity verification and relationship mapping, thereby reducing the risk of energy dispatch instruction errors caused by misjudgment of energy dispatch requirement signals or differences in working state, and improving the operational safety of photovoltaic energy storage system energy dispatch.

[0129] In practical implementation, the scheduling strategy mentioned above includes:

[0130] When there is sufficient sunlight, after the load power supply is balanced, the photovoltaic system's power generation will first supply the energy storage battery for charging, and then supply the grid.

[0131] When sunlight is insufficient, the photovoltaic system prioritizes power generation to supply user loads, and the energy storage system prioritizes power discharge to supply user loads. If power generation and discharge cannot meet the user load demand, then the grid is the next priority to purchase electricity.

[0132] When the power grid fails, the photovoltaic energy storage system switches to off-grid mode, prioritizing power supply to user loads through photovoltaic power generation and energy storage discharge;

[0133] During off-peak electricity prices, electricity is purchased from the grid at a low price to charge the energy storage system;

[0134] During peak electricity prices, the energy storage system supplies power to user loads and discharges to the grid for sale at higher prices.

[0135] As can be seen, the above embodiments define the conditions of the scheduling strategy, comprehensively consider the energy scheduling needs under different conditions such as illumination, grid status, load status, and electricity price, and can meet the energy scheduling needs under different conditions. This enables data time-series analysis based on different working conditions, accurately generates energy scheduling instructions, and improves the accuracy and operational safety of intelligent energy scheduling operation of photovoltaic energy storage systems.

[0136] The unique technical advantage of this application lies in the fact that by acquiring the working status data of the photovoltaic energy storage system at historical time points through sensors, a state prediction model is trained. Based on the data at the previous time point, the scheduling demand at the current time point is predicted and the energy scheduling command is further determined. This enables the generation of accurate energy scheduling commands based on the time sequence analysis of the working status, improving the accuracy and operational safety of intelligent control of energy scheduling in the photovoltaic energy storage system. It has high application value in the field of energy storage system control technology.

[0137] Please see Figure 4 According to one aspect of this application, an energy dispatching platform system for a photovoltaic energy storage system is provided. The platform system includes: an acquisition module for acquiring operating status data of the photovoltaic energy storage system at multiple historical time points through sensors; a training module for training a state prediction model based on the operating status data corresponding to each of the historical time points; a prediction module for predicting the predicted dispatching demand at the current time point based on the state prediction model and the operating status data at the previous time point; and a determination module for determining the energy dispatching instruction corresponding to the photovoltaic energy storage system based on the predicted dispatching demand in response to the energy dispatching demand signal at the current time point.

[0138] Based on any embodiment of the system in this application, the system of this application further includes: a training data acquisition module, configured to determine the state feature parameters corresponding to each working state data based on state recognition rules; sort each state feature parameter from early to late based on the historical time points to obtain a parameter time series; and use the parameter time series as a training dataset to train a preset long short-term memory neural network to obtain a state prediction model.

[0139] Based on any embodiment of the system in this application, the training data acquisition module in the system of this application includes working status data such as light intensity, photovoltaic power generation, energy storage battery voltage, energy storage battery current, energy storage battery temperature, energy storage battery status, and electricity price data; the status characteristic parameters include light intensity, photovoltaic power generation, energy storage capacity, load demand, real-time electricity price, load type, and electricity price type; the grid load type includes power generation greater than the sum of energy storage capacity demand and user load demand, power generation less than user load demand, and power generation greater than or equal to user load demand; the electricity price type includes off-peak electricity price and peak electricity price.

[0140] Based on any embodiment of the system in this application, the system further includes: a model training module, configured to, for any two adjacent state feature parameters within the parameter time series, calculate the difference between the light intensity of the two adjacent state feature parameters to obtain a light intensity difference; calculate the difference between the photovoltaic power generation of the two adjacent state feature parameters to obtain a power generation difference; calculate the difference between the energy storage capacity of the two adjacent state feature parameters to obtain an energy storage difference; calculate the difference between the load demand of the two adjacent state feature parameters to obtain a load difference; calculate the difference between the real-time electricity price of the two adjacent state feature parameters to obtain an electricity price difference; calculate the weighted sum of the light intensity difference, the power generation difference, the energy storage difference, the load difference, and the electricity price difference to obtain a changing feature parameter; calculate the ratio of the changing feature parameter to the time difference between the historical time points corresponding to the two state feature parameters to obtain the correlation feature between the two state feature parameters; and concatenate the parameter time series and each corresponding correlation feature into a fusion feature, input it into a preset long short-term memory neural network for training, to obtain a state prediction model.

[0141] Based on any embodiment of the system in this application, in the model training module of the system in this application, the loss function of the long short-term memory neural network is set as the product of a first cross-entropy, a second cross-entropy, and a third cross-entropy. The first cross-entropy is the difference between the predicted output and the label corresponding to the parameter time series; the second cross-entropy is the difference between the output association feature and the actual association feature; the third cross-entropy is the difference between the predicted output and the scheduling demand representation of the output association feature and the real label corresponding to the actual scheduling demand; the output association feature is the association feature between the predicted output and the predicted output of the previous time step; and the actual association feature is the actual association feature between the corresponding parameters of the parameter time series.

[0142] Based on any embodiment of the system in this application, the sensor in the acquisition module of the system in this application includes a light sensor, a temperature sensor, a battery status sensor, and a system status sensor. The light sensor is used to detect the ambient light intensity, the temperature sensor is used to detect the operating temperature of the photovoltaic energy storage system, the battery status sensor is used to detect the voltage, current, and state of charge of the energy storage battery in the energy storage system, and the system status sensor is used to identify the load type and electricity price type based on the operating status data.

[0143] Based on any embodiment of the system in this application, the system of this application further includes: a demand prediction module, configured to acquire working status data of at least the previous two time points of the current time point and determine them as historical status data; calculate the correlation features between the two historical status data; and input the historical status data and the corresponding correlation features into the status prediction model to obtain the predicted scheduling demand of the current time point.

[0144] Based on any embodiment of the system in this application, the system further includes: an instruction determination module, configured to acquire current operating state data and corresponding scheduling demand parameters corresponding to the energy scheduling demand signal; determine reference state features corresponding to the predicted scheduling demand in a preset database corresponding to the scheduling demand parameters; calculate the feature similarity between the state features of the current operating state data and the reference state features; determine whether the feature similarity is greater than a preset similarity threshold; if not, determine the energy scheduling instruction corresponding to the photovoltaic energy storage system through actual user interface operation of the scheduling platform; if yes, determine the energy scheduling instruction of the photovoltaic energy storage system based on the predicted scheduling demand and the preset correspondence between operating states and scheduling strategies.

[0145] Based on any embodiment of the system in this application, the scheduling strategy in the instruction determination module of the system in this application includes: when there is sufficient sunlight, after the load power supply is balanced, the photovoltaic system generates electricity to charge the energy storage battery first, and then supplies it to the grid; when there is insufficient sunlight, the photovoltaic system generates electricity to supply the user load first, and the energy storage system discharges to supply the user load first. If the generation and discharge do not meet the user load demand, the grid is selected as the second priority for power purchase; when the grid is shut down, the photovoltaic energy storage system switches to off-grid mode and supplies the user load first through photovoltaic generation and energy storage discharge; during off-peak electricity prices, the energy storage system is charged by purchasing electricity from the grid at a low price; during peak electricity prices, the energy storage system supplies the user load and discharges to the grid for high-price electricity sales.

[0146] Another embodiment of this application provides an energy dispatching device for a photovoltaic energy storage system. This energy dispatching device includes a processor, a computer-readable storage medium, a memory, and a network interface connected via a system bus. The computer-readable, non-volatile storage medium of the energy dispatching device stores an operating system, a database, and computer-readable instructions. The database may store information sequences. When the computer-readable instructions are executed by the processor, the processor can implement an energy dispatching method for the photovoltaic energy storage system.

[0147] The processor of the energy dispatching device in this photovoltaic energy storage system provides computing and control capabilities to support the operation of the entire photovoltaic energy storage system's energy dispatching device. The memory of the energy dispatching device can store computer-readable instructions, which, when executed by the processor, cause the processor to perform the energy dispatching method of the photovoltaic energy storage system described in this application. The network interface of the energy dispatching device is used for communication with terminals.

[0148] In this embodiment, the processor is used to execute... Figure 4 The system defines the specific functions of each module, and the memory stores the program code and various data required to execute these modules or submodules. The network interface is used to enable data transmission between user terminals or servers.

[0149] The non-volatile readable storage medium in this embodiment stores the program code and data required to execute all modules in the energy dispatch platform system of the photovoltaic energy storage system of this application. The server can call the program code and data of the server to execute the functions of all modules.

[0150] This application also provides a non-volatile readable storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the energy dispatching method of the photovoltaic energy storage system according to any embodiment of this application.

[0151] This application also provides a computer program product, including a computer program / instructions that, when executed by one or more processors, implement the steps of the method described in any embodiment of this application.

Claims

1. An energy scheduling method for a photovoltaic energy storage system, characterized in that, The method comprises the following steps: obtaining working state data of a photovoltaic energy storage system at a plurality of historical time points; training a state prediction model based on the working state data corresponding to each of the historical time points; predicting a predicted scheduling requirement at a current time point based on the working state data at a previous time point based on the state prediction model; determining an energy scheduling instruction corresponding to the photovoltaic energy storage system according to the predicted scheduling requirement in response to an energy scheduling requirement signal at the current time point; training a state prediction model based on the working state data corresponding to each of the historical time points comprises: determining a state feature parameter corresponding to each of the working state data based on a state recognition rule; sorting each of the state feature parameters from early to late based on the historical time points to obtain a parameter time sequence; training a preset long short-term memory neural network using the parameter time sequence as a training data set to obtain the state prediction model; the working state data comprises illumination data, photovoltaic power generation power, energy storage battery voltage, energy storage battery current, energy storage battery temperature, energy storage battery state and electricity price data; the state feature parameter comprises illumination intensity, photovoltaic power generation capacity, energy storage capacity, load demand, real-time electricity price, load type and electricity price type; the load type comprises photovoltaic power generation capacity greater than the sum of energy storage capacity demand and user load demand, photovoltaic power generation capacity less than user load demand and photovoltaic power generation capacity greater than or equal to user load demand; the electricity price type comprises off-peak electricity price and peak electricity price; training a preset long short-term memory neural network using the parameter time sequence as a training data set to obtain the state prediction model comprises: calculating a difference value between the illumination intensity of any two adjacent state feature parameters in the parameter time sequence to obtain an illumination difference value; calculating a difference value between the photovoltaic power generation capacity of any two adjacent state feature parameters to obtain a power generation difference value; calculating a difference value between the energy storage capacity of any two adjacent state feature parameters to obtain an energy storage difference value; calculating a difference value between the load demand of any two adjacent state feature parameters to obtain a load difference value; calculating a difference value between the real-time electricity price of any two adjacent state feature parameters to obtain an electricity price difference value; calculating a weighted sum of the illumination difference value, the power generation difference value, the energy storage difference value, the load difference value and the electricity price difference value to obtain a change feature parameter; calculating a ratio of the change feature parameter to a time difference value between the historical time points corresponding to the two state feature parameters to obtain a correlation feature between the two state feature parameters; concatenating the parameter time sequence and each of the corresponding correlation features into a fusion feature and inputting the fusion feature into a preset long short-term memory neural network for training to obtain the state prediction model; The loss function of the long short-term memory neural network is set as a product of a first cross entropy, a second cross entropy and a third cross entropy, the first cross entropy is a difference between a predicted output and a label corresponding to the parameter time sequence; the second cross entropy is a difference between an output correlation feature and an actual correlation feature; the third cross entropy is a difference between a predicted output and a real label corresponding to a scheduling demand representation of the output correlation feature; the output correlation feature is a correlation feature between the predicted output and a predicted output at a previous time step; and the actual correlation feature is an actual correlation feature between corresponding parameters of the parameter time sequence.

2. The energy scheduling method of the photovoltaic energy storage system according to claim 1, wherein, The working state data of the photovoltaic energy storage system at the plurality of historical time points is obtained by a sensor, the sensor includes an illumination sensor, a temperature sensor, a battery state sensor and a system state sensor, the illumination sensor is used to detect ambient light intensity, the temperature sensor is used to detect the working temperature of the photovoltaic energy storage system, the battery state sensor is used to detect the voltage, current and state of charge of the energy storage battery in the energy storage system, and the system state sensor is used to identify the load type and the electricity price type according to the working state data.

3. The method of energy scheduling for a photovoltaic energy storage system of claim 1, wherein, Based on the state prediction model, the predicted scheduling demand at the current time point is predicted according to the working state data at the previous time point, including: obtaining the working state data of at least the previous two time points at the current time point, and determining the historical state data; calculating the correlation feature between the two historical state data; inputting the historical state data and the corresponding correlation feature into the state prediction model to obtain the output predicted scheduling demand at the current time point.

4. The energy scheduling method of the photovoltaic energy storage system according to claim 3, characterized in that, According to the predicted scheduling demand, the energy scheduling instruction corresponding to the photovoltaic energy storage system is determined, including: obtaining the current working state data corresponding to the energy scheduling demand signal and the corresponding scheduling demand parameter; determining the reference state feature corresponding to the predicted scheduling demand in the preset database corresponding to the scheduling demand parameter; calculating the feature similarity between the state feature of the current working state data and the reference state feature; determining whether the feature similarity is greater than a preset similarity threshold; if not, the actual user interface operation of the scheduling platform is used to determine the energy scheduling instruction corresponding to the photovoltaic energy storage system; if yes, the energy scheduling instruction of the photovoltaic energy storage system is determined based on the corresponding relationship between the preset working state and the scheduling strategy according to the predicted scheduling demand.

5. The energy scheduling method of the photovoltaic energy storage system according to claim 4, wherein, The scheduling strategy includes: when the light is sufficient, the photovoltaic system generates electricity to preferentially charge the energy storage battery after balancing the load power supply, and secondarily preferentially supply the power grid; when the light is insufficient, the photovoltaic system generates electricity to preferentially supply the user load, the energy storage system preferentially discharges to supply the user load, and if the power generation and discharge do not meet the user load demand, the power grid is secondarily preferentially selected to purchase electricity; when the power grid is powered off, the photovoltaic energy storage system switches to an off-grid mode, and the photovoltaic power generation and energy storage discharge are used to preferentially supply the user load; when the valley electricity price is low, the energy storage system is charged by purchasing low-price electricity from the power grid. During peak electricity price, the energy storage system supplies user load and discharges to supply the grid for high price electricity sale.

6. An energy scheduling platform system for a photovoltaic energy storage system, characterized by, The system is used to execute the energy scheduling method of the photovoltaic energy storage system in any one of claims 1-5, and the system comprises: An acquisition module is configured to acquire working state data of the photovoltaic energy storage system at a plurality of historical time points through a sensor; A training module is configured to train a state prediction model based on the working state data corresponding to each of the historical time points; A prediction module is configured to predict a predicted scheduling demand at a current time point based on the working state data at a previous time point according to the state prediction model; A determination module is configured to determine an energy scheduling instruction corresponding to the photovoltaic energy storage system according to the predicted scheduling demand in response to an energy scheduling demand signal at the current time point.

Images