A method, apparatus, medium and system for secure grid connection of a distributed photovoltaic energy storage system

By constructing a secure grid connection estimation model and a comprehensive evaluation model based on a long short-term memory neural network, and combining them with a 5G communication module, the problem of secure grid connection status monitoring of distributed optical energy storage systems was solved. This enabled efficient and secure grid connection and real-time data transmission of distributed optical energy storage systems, improving the system's security and applicability.

CN116707006BActive Publication Date: 2026-11-17STATE GRID HUNAN ELECTRIC POWER COMPANY LIMITED +2
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202310735699.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-06-20
Publication Date
2026-11-17
Estimated Expiration
2043-06-20

AI Technical Summary

Technical Problem

In existing technologies, it is difficult to monitor the safe grid connection status of distributed photovoltaic energy storage systems. It is impossible to effectively monitor parameters such as frequency and voltage at the grid connection point, resulting in poor safety and transmission practicality, and making it difficult to guarantee the stable and safe grid connection of the system.

Method used

A safe grid connection estimation model based on long short-term memory neural network is adopted and combined with Adam optimization algorithm to construct a safe grid connection estimation model for distributed photovoltaic storage system. The safety is evaluated through a comprehensive evaluation model, and the frequency, voltage, current and power parameters of the grid connection point are monitored. Data is remotely transmitted to the safety monitoring system center using 5G communication module.

Benefits of technology

It improves the security and real-time monitoring of distributed photovoltaic storage systems, reduces engineering costs and operation and maintenance costs, enables flexible real-time data monitoring and long-distance data transmission of distributed photovoltaic storage systems, and enhances the security and applicability of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116707006B_ABST
    Figure CN116707006B_ABST
Patent Text Reader

Abstract

This invention discloses a method, apparatus, medium, and system for secure grid connection of a distributed photovoltaic (PV) storage system. The method includes: acquiring historical parameters of the PV storage system's grid connection points at different time points; inputting the historical parameters into a pre-constructed PV storage system secure grid connection estimation model to obtain estimated grid connection status values ​​at corresponding time points, which are used as preset thresholds for secure grid connection; acquiring real-time status parameters of the grid connection points and comparing them with the corresponding preset thresholds for secure grid connection; when any parameter in the real-time status parameters is not within the corresponding preset threshold, evaluating the secure grid connection status of the PV storage system using a preset comprehensive evaluation model to obtain an evaluation score; when the evaluation score is lower than a preset evaluation score threshold, determining that there is a security risk in the current grid connection of the PV storage system. This invention can effectively improve the security of grid connection for distributed PV storage systems.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention mainly relates to the field of new energy technology, specifically to a method, device, medium, and system for the safe grid connection of a distributed photovoltaic energy storage system. Background Technology

[0002] With the continuous development of the new energy industry, the proportion of distributed photovoltaic (PV) and energy storage (ESS) systems connected to large power grids is constantly increasing, and their applications are becoming more widespread. Currently, the large-scale grid connection of ESS systems puts pressure on the safe and stable operation of the power grid. At the same time, due to the dispersed deployment locations, complex operating environments, large number, small scale, and significant differences of ESS systems, unified management is difficult. This results in significant challenges in monitoring the safe grid connection status of ESS systems, and relatively lagging operational management, leading to an increasing number of accidents related to the safe grid connection of ESS systems.

[0003] One such energy storage system comprises a battery array of tens of thousands of lithium batteries. Current monitoring systems primarily monitor parameters such as voltage, current, temperature, and humidity of the battery array in the distributed photovoltaic-energy storage system in real time. They do not effectively monitor parameters such as frequency and voltage at the grid connection point of the distributed photovoltaic-energy storage system, or they only monitor one or two parameters. Therefore, existing methods for monitoring the safe grid connection status of distributed photovoltaic-energy storage systems are inadequate, with poor security and practicality in transmission, failing to meet the needs of the increasingly sophisticated distributed photovoltaic-energy storage systems. Consequently, it is difficult to guarantee the stable and safe grid connection of distributed photovoltaic-energy storage systems. Summary of the Invention

[0004] The technical problem to be solved by this invention is: in view of the technical problems existing in the prior art, this invention provides a method, device, medium and system for safely connecting a distributed optical energy storage system to the grid, which can effectively improve the grid connection security of the distributed optical energy storage system.

[0005] To solve the above-mentioned technical problems, the technical solution proposed by this invention is as follows:

[0006] A method for secure grid connection of a distributed optical energy storage system includes the following steps:

[0007] S1. Obtain historical parameters of distributed photovoltaic energy storage system grid connection points at different time points and construct a historical feature dataset; the historical parameters include frequency, voltage, current and power;

[0008] S2. Input the historical parameters of the distributed photovoltaic and energy storage system grid connection points at each time point into the pre-built distributed photovoltaic and energy storage system safe grid connection estimation model to obtain the estimated value of the safe grid connection status of the distributed photovoltaic and energy storage system at the corresponding time point.

[0009] S3. Use the estimated safe grid connection status of the distributed photovoltaic storage system at each time point as the preset threshold for safe grid connection of the distributed photovoltaic storage system.

[0010] S4. Obtain the real-time status parameters of the grid connection point of the distributed photovoltaic storage system and compare them with the corresponding safe grid connection preset thresholds; when all parameters in the real-time status parameters are within the corresponding safe grid connection preset thresholds, grid connection is performed; when any parameter in the real-time status parameters is not within the corresponding safe grid connection preset thresholds, proceed to step S5; where the real-time status parameters include the grid connection point frequency, voltage, current and power.

[0011] S5. Use a preset comprehensive evaluation model to evaluate the safe grid connection status of the distributed photovoltaic storage system and obtain an evaluation score. When the evaluation score is lower than the preset evaluation score threshold, it is determined that there is a safety risk in the current grid connection of the distributed photovoltaic storage system. If the evaluation score is within the preset evaluation score threshold, then grid connection is carried out.

[0012] Preferably, in step S2, the construction process of the corresponding distributed photovoltaic storage system safe grid connection estimation model is as follows: an initial distributed photovoltaic storage system safe grid connection estimation model is constructed based on a long short-term memory neural network, and then the Adam optimization algorithm is used to obtain the optimized distributed photovoltaic storage system safe grid connection estimation model.

[0013] Preferably, the loss function of the distributed photovoltaic-storage system safe grid connection estimation model adopts the mean squared error loss function formula:

[0014]

[0015] Where: y i This is the current real value of the safe operating status of the distributed optical energy storage system; It is the estimated safe operating status of the distributed optical energy storage system at the current moment; M is the number of samples.

[0016] Preferably, the Adam optimization algorithm is as follows:

[0017] s t =ρ1s t-1 +(1-ρ1)g;

[0018] r t =ρ2r t-1 +(1-ρ2)g 2 ;

[0019]

[0020]

[0021]

[0022] Where: s tLet represent the smoothed first moment after t iterations; t represents the number of iterations; ρ1 and ρ2 represent the decay constants; g represents the gradient value; r t θ represents the smoothed second moment after t iterations; t θ t+1 The parameters are updated for the model at iterations t and t+1; μ represents the learning rate for iteration t; ε = 10. -8 Parameters set to avoid division by zero; ρ1 t and ρ2 t It is the t-th power of the decay constant.

[0023] Preferably, the preset comprehensive evaluation model is as follows:

[0024]

[0025] Where X represents the evaluation score; m represents the number of parameter types; S 0i S represents the preset safe grid connection threshold corresponding to the i-th parameter; i It is the detection data corresponding to the i-th parameter when any detection parameter value exceeds its corresponding safe grid connection preset threshold.

[0026] This invention also discloses a safe grid-connected control device for a distributed photovoltaic energy storage system, comprising a data metering module, a data storage module, an amplifier module, a central chip module, a filtering module, and a communication module. The data metering module, amplifier module, central chip module, filtering module, and communication module are sequentially connected, and the data storage module is connected to the data metering module. The data metering module measures the frequency, voltage, current, and power parameters of the distributed photovoltaic energy storage system's grid connection point and converts them into digital signals for transmission to the amplifier module. The amplifier module amplifies and converts the received parameter signals and transmits the amplified signals to the central chip module for data calculation and processing, then transmits the processed data to the filtering module. The filtering module filters the output signal of the central chip module before transmitting it to the communication module. The communication module remotely transmits the output signal of the central chip module to the safety monitoring system center. The data storage module is configured to store the grid connection point parameters collected by the data metering module.

[0027] Preferably, the amplifier module is a signal amplification module based on LM324.

[0028] Preferably, the output signal of the central chip module is transmitted to the filtering module via serial communication; the serial communication is an RS485-based interface.

[0029] The present invention also discloses a computer-readable storage medium having a computer program stored thereon, the computer program performing the steps of the method described above when run by a processor.

[0030] The present invention further discloses a safe grid-connected system for a distributed optical storage system, comprising a memory and a processor connected to each other, wherein the memory stores a computer program, and the computer program executes the steps of the method described above when run by the processor.

[0031] Compared with the prior art, the advantages of the present invention are as follows:

[0032] 1. The distributed photovoltaic energy storage system safe grid connection estimation model of the present invention is based on a deep construction of a long short-term memory neural network and uses the Adam optimization algorithm for model optimization. It can adapt the estimation model to parameter data conditions at different time points using a relatively small amount of historical data, which can reduce the training difficulty and cost of the estimation model, improve the applicability of the estimation model, and improve the accuracy of the estimation model. The present invention also adopts a comprehensive evaluation model to evaluate the safe grid connection status of the distributed photovoltaic energy storage system, thereby improving the safety of the distributed photovoltaic energy storage system grid connection.

[0033] 2. This invention monitors the frequency, voltage, current and power parameters at the grid connection point of a distributed photovoltaic energy storage system, expanding the types of parameters that traditional distributed photovoltaic energy storage monitoring systems can detect, and effectively improving the security of grid connection of distributed photovoltaic energy storage systems.

[0034] 3. This invention overcomes the shortcomings of traditional distributed optical storage systems, such as poor real-time monitoring, poor security, weak long-distance information transmission capability, limited data monitoring capacity, and high installation costs. It can flexibly and conveniently perform real-time data monitoring on the distributed optical storage system connected to the grid, and can also transmit data to the security monitoring system center over long distances in a timely manner. The overall system has high data transmission security, requires no wiring, has strong scalability, and greatly reduces engineering costs and operation and maintenance costs. Attached Figure Description

[0035] Figure 1 The flowchart illustrates the safe grid connection method for the distributed optical energy storage system of the present invention in an embodiment.

[0036] Figure 2 This is a flowchart of the long short-term memory neural network structure in this invention.

[0037] Figure 3 This is a structural block diagram of the distributed photovoltaic energy storage system safe grid connection control device of the present invention in an embodiment. Detailed Implementation

[0038] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0039] like Figure 1 As shown, the distributed optical energy storage system secure grid connection method of this invention includes the following steps:

[0040] S1. Obtain historical parameters of the grid-connected points of the distributed photovoltaic energy storage system at different time points and construct a historical feature dataset; the historical parameters include the grid-connected point frequency, voltage, current and power;

[0041] S2. Input the historical feature dataset from step S1 into the distributed photovoltaic-storage system safety grid connection estimation model constructed based on long short-term memory neural network for training to obtain a pre-estimation model; then use the Adam optimization algorithm to obtain the optimized safety grid connection estimation model; then train the optimized safety grid connection estimation model with the historical parameters of the distributed photovoltaic-storage system safety grid connection status at different time points to adjust the parameters of its fully connected output layer. Training stops when the accuracy of the safety grid connection estimation model reaches the preset accuracy, and the trained distributed photovoltaic-storage system safety grid connection estimation model is obtained, outputting the estimated value of the distributed photovoltaic-storage system safety grid connection status at the corresponding time point;

[0042] The safe grid connection estimation model is based on a long short-term memory neural network, including an input gate, a forget gate, and an output gate, as shown in the following formula:

[0043] i t =σ(W i ·[h t-1 ,x t ]+b i );

[0044] f t =σ(W f ·[h t-1 ,x t ]+b f );

[0045] O t =σ(W O [h t-1 ,x t ]+b O ;

[0046]

[0047]

[0048] h t =O t *tanh(C t );

[0049] In the above formula: σ represents the sigmoid activation function; i t It is the input gate, which uses the sigmoid function to control which information needs to be saved to the internal cell state; f tIt's a forget gate, which selectively filters past state information using the sigmoid function; O t It is an output gate that uses the sigmoid function to control the output of internal state information to the external state. It is the state of the newly generated candidate unit; C t It is the unit state; h t-1 ,h t The values ​​represent the previous and current outputs of the unit; W and b represent the weight matrices; x t This indicates the input at the current time;

[0050] The loss function of the distributed photovoltaic storage system safe grid connection estimation model adopts the mean square error loss function formula:

[0051]

[0052] In the above formula: y i This is the current real value of the safe operating status of the distributed optical energy storage system; It is the estimated safe operating status of the distributed optical energy storage system at the current moment; M is the number of samples.

[0053] Among them, the Adam optimization algorithm used to optimize the estimation model for the safe grid connection of distributed photovoltaic storage systems is as follows:

[0054] s t =ρ1s t-1 +(1-ρ1)g;

[0055] r t =ρ2r t-1 +(1-ρ2)g 2 ;

[0056]

[0057]

[0058]

[0059] In the above formula: s t Let represent the smoothed first moment after t iterations; t represents the number of iterations; ρ1 and ρ2 represent the decay constants; g represents the gradient value; r t θ represents the smoothed second moment after t iterations; t θ t+1 The parameters are updated for the model at iterations t and t+1; μ represents the learning rate for iteration t; ε = 10. -8 Parameters set to avoid division by zero; ρ1 t and ρ2 t It is the t-th power of the decay constant.

[0060] S3. Use the estimated safe grid connection status of the distributed optical storage system at each time point as the preset threshold for safe grid connection of the distributed optical storage system.

[0061] S4. Obtain the real-time status parameters of the grid connection point of the distributed photovoltaic storage system, and compare the real-time status parameters of the grid connection point with its corresponding safe grid connection preset threshold; when all parameters in the real-time status parameters are within the corresponding safe grid connection preset threshold, grid connection can be safely carried out; when any one of the grid connection point frequency, voltage, current and power parameters in the real-time status parameters exceeds its corresponding preset safe grid connection threshold, proceed to step S5.

[0062] S5. Use a comprehensive evaluation model to evaluate the safe grid connection status of the distributed photovoltaic energy storage system and obtain an evaluation score; compare the evaluation score with a preset evaluation score threshold; when the evaluation score is lower than the preset evaluation score threshold, it is determined that there is a safety risk in the current grid connection of the distributed photovoltaic energy storage system. At this time, a remote control command is issued to disconnect the grid-connected distributed photovoltaic energy storage system and an alarm is triggered; if the evaluation score is within the preset evaluation score threshold, grid connection can be safely carried out.

[0063] The comprehensive evaluation model is as follows:

[0064]

[0065] Where X represents the evaluation score; m represents the number of parameter types including the grid connection point frequency detection data, voltage detection data, current detection data, and frequency monitoring data; S 0i S represents the preset safe grid connection threshold corresponding to the i-th parameter data; i It is the detection data corresponding to the i-th parameter type when any detection data parameter value exceeds its corresponding safe grid connection preset threshold parameter.

[0066] The distributed photovoltaic (PV) and energy storage (EPS) system safety grid connection estimation model of this invention is based on a deep Long Short-Term Memory (LSTM) neural network and optimized using the Adam optimization algorithm. Utilizing relatively small-scale historical data, the estimation model can adapt to parameter data conditions at different time points, reducing training difficulty and cost, expanding its applicability, and improving accuracy. This invention also employs a comprehensive evaluation model to assess the safety grid connection status of the EPS system, thereby improving the safety of EPS grid connection. This invention monitors the frequency, voltage, current, and power parameters at the EPS grid connection point, expanding the types of parameters that traditional EPS monitoring systems can detect and effectively improving the safety of EPS grid connection.

[0067] like Figure 3As shown, this embodiment of the invention also provides a safe grid-connected control device for a distributed photovoltaic energy storage system, specifically including a data metering module, a data storage module, an amplifier module, a central chip module, a filtering module, and a 5G communication module. The data metering module, amplifier module, central chip module, filtering module, and communication module are connected sequentially, and the data storage module is connected to the data metering module. The data metering module measures the parameters of frequency, voltage, current, and power at the grid connection point of the distributed photovoltaic energy storage system and converts them into digital signals, which are then transmitted to the amplifier module. The amplifier module amplifies and converts the received parameter data signals and transmits the amplified signals to the central chip module for data calculation and processing. The processed data is then transmitted to the filtering module via serial communication. The filtering module filters the output signal of the central chip module and transmits it to the 5G communication module. The 5G communication module remotely transmits the data to the safety monitoring system center via the 5G communication network. The data storage module is configured to store the grid connection point parameters collected by the data metering module. LM5007 and LM5017 step-down chips are used to power the entire grid-connected control device. The amplifier module is a signal amplification module based on LM324. The serial communication is based on an RS485 interface. In the event of a grid connection anomaly, the safety monitoring system center sends a remote control command to the Bluetooth grid connection circuit breaker to disconnect the grid-connected distributed optical storage system and trigger an alarm.

[0068] The distributed photovoltaic storage system safe grid connection control device of the present invention is used to execute the grid connection method described above. It also has the advantages described above, and at the same time, it has a simple structure and is easy to implement.

[0069] This invention overcomes the shortcomings of traditional distributed optical storage systems, such as poor real-time monitoring, poor security, weak long-distance information transmission capability, limited data monitoring capacity, and high installation costs. It can flexibly and conveniently perform real-time data monitoring on the distributed optical storage system connected to the grid, and can also transmit data to the security monitoring system center over long distances in a timely manner. The overall system has high data transmission security, requires no wiring, has strong scalability, and greatly reduces engineering costs and operation and maintenance costs.

[0070] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when run by a processor, executes the steps of the method described above. This invention further provides a distributed optical storage system with secure grid connection, including an interconnected memory and a processor. The memory stores a computer program, which, when run by a processor, executes the steps of the method described above. The medium and system of this invention correspond to the methods described above and also possess the advantages described above.

[0071] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A method for secure grid connection of a distributed photovoltaic energy storage system, characterized in that, Including the following steps: S1. Obtain historical parameters of distributed photovoltaic energy storage system grid connection points at different time points and construct a historical feature dataset; the historical parameters include frequency, voltage, current and power; S2. Input the historical parameters of the distributed photovoltaic and energy storage system grid connection points at each time point into the pre-built distributed photovoltaic and energy storage system safe grid connection estimation model to obtain the estimated value of the safe grid connection status of the distributed photovoltaic and energy storage system at the corresponding time point. S3. Use the estimated safe grid connection status of the distributed photovoltaic storage system at each time point as the preset threshold for safe grid connection of the distributed photovoltaic storage system. S4. Obtain the real-time status parameters of the grid connection point of the distributed optical storage system and compare them with the corresponding safe grid connection preset threshold. If all parameters in the real-time status parameters are within the corresponding safe grid connection preset threshold, then grid connection is performed; if any parameter in the real-time status parameters is not within the corresponding safe grid connection preset threshold, then proceed to step S5. The real-time status parameters include grid connection point frequency, voltage, current, and power; S5. Use a preset comprehensive evaluation model to evaluate the safe grid connection status of the distributed photovoltaic storage system and obtain an evaluation score; when the evaluation score is lower than the preset evaluation score threshold, it is determined that there is a safety risk in the current grid connection of the distributed photovoltaic storage system; if the evaluation score is within the preset evaluation score threshold, then grid connection is carried out. In step S2, the construction process of the corresponding distributed photovoltaic storage system safe grid connection estimation model is as follows: the initial distributed photovoltaic storage system safe grid connection estimation model is constructed based on the long short-term memory neural network, and then the Adam optimization algorithm is used to obtain the optimized distributed photovoltaic storage system safe grid connection estimation model. The loss function of the distributed photovoltaic-storage system safe grid connection estimation model adopts the mean squared error loss function formula: ; in: This is the current real value of the safe operating status of the distributed optical energy storage system; It is the estimated safe operating status of the distributed optical energy storage system at the current moment; M is the number of samples.

2. The method for secure grid connection of a distributed photovoltaic storage system according to claim 1, characterized in that, The Adam optimization algorithm is as follows: ; ; ; ; ; in: Let represent the smoothed first moment after t iterations; t represents the number of iterations. and Indicates the attenuation constant; Indicates the gradient value; Let represent the smoothed second moment after t iterations; , Update the parameters for the model at the t-th and t+1-th iterations; This represents the learning rate after t iterations. =10 -8 A parameter set to avoid division by zero; and It is the t-th power of the decay constant.

3. The method for secure grid connection of a distributed photovoltaic energy storage system according to claim 1 or 2, characterized in that, The pre-set comprehensive evaluation model is as follows: in, Indicates the evaluation score; Indicates the number of parameter types; This represents the preset threshold for safe grid connection corresponding to the i-th parameter; It is the detection data corresponding to the i-th parameter when any detection parameter value exceeds its corresponding safe grid connection preset threshold.

4. A distributed photovoltaic energy storage system secure grid connection control device, used to perform the steps of the method as described in any one of claims 1 to 3, characterized in that, The system includes a data metering module, a data storage module, an amplifier module, a central chip module, a filtering module, and a communication module. These modules are sequentially connected, and the data storage module is connected to the data metering module. The data metering module measures the frequency, voltage, current, and power parameters of the distributed photovoltaic energy storage system's grid connection point and converts them into digital signals for transmission to the amplifier module. The amplifier module amplifies and converts the received parameter signals, then transmits the amplified signals to the central chip module for data processing, and subsequently transmits the processed data to the filtering module. The filtering module filters the output signal from the central chip module before transmitting it to the communication module. The communication module remotely transmits the output signal from the central chip module to the safety monitoring system center. The data storage module is configured to store the grid connection point parameters collected by the data metering module.

5. The distributed photovoltaic energy storage system secure grid connection control device according to claim 4, characterized in that, The amplifier module is a signal amplification module based on LM324.

6. The distributed photovoltaic energy storage system safe grid connection control device according to claim 4 or 5, characterized in that, The output signal of the central chip module is transmitted to the filtering module via serial communication; the serial communication is an RS485-based interface.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program, when run by a processor, performs the steps of the method as described in any one of claims 1 to 3.

8. A distributed optical energy storage system with secure grid connection, comprising interconnected memory and processor, wherein the memory stores a computer program, characterized in that, The computer program, when run by a processor, performs the steps of the method as described in any one of claims 1 to 3.

Citation Information

Patent Citations

  • Method, device and system for analyzing and evaluating distributed power supply access to power grid

    CN105956939A

  • Comprehensive evaluation method and device for performance of electricity utilization information acquisition equipment

    CN115719173A