Photovoltaic energy storage power station remote monitoring Internet of Things control method and system

Abstract

The present invention relates to the field of photovoltaic power station monitoring technology, and in particular to a method and system for remote monitoring, Internet of Things, and physical control of photovoltaic energy storage power stations. The method collects energy storage battery status data, grid power data, and relay status data through a data acquisition device and sends the data to the cloud through the Internet of Things. The data and grid power data are obtained from the cloud, and the battery energy storage status and grid power data are analyzed using a power monitoring model. The relay status data, battery status data, and grid power data are integrated through a fusion monitoring model to predict the switching status of the relay and automatically control the normal connection or disconnection of the energy storage battery with the grid and / or photovoltaic power generation unit. The present invention combines the existing composition structure of the energy storage power station with a machine learning algorithm model to monitor the power of the grid, predict changes in grid power, automatically control relay operation, and utilize the charging and discharging of the energy storage battery to regulate grid power.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic energy storage power station monitoring, and in particular to a photovoltaic energy storage power station remote monitoring, Internet of Things, and Control method and system. Background Art

[0002] In existing photovoltaic power plants, photovoltaic panels are connected in series to achieve a certain voltage. After multiple photovoltaic strings are concentrated and combined through a combiner box, they are connected to the inverter and then stepped up by a transformer to be connected to the medium and high voltage power grid. For photovoltaic power plants with energy storage units, the energy storage units are connected to the combiner box through an inverter or a charge and discharge module, so that the current output by the photovoltaic panels can be stored in the energy storage battery pack to ensure the stability of the power supply of the power station. The specific composition of the photovoltaic power station can be referred to in the attached Figure 3 Active power is the power delivered to the grid by a PV power station, which directly impacts grid stability and the supply-demand balance. Reactive power, on the other hand, significantly impacts the grid's voltage stability and power factor. To effectively control the active and reactive power delivered to the grid by a PV power station, it's essential to constantly monitor the grid's power status and implement appropriate regulatory measures to better adapt to dynamic grid changes.

[0003] With the continued development of smart grid technology, the interaction between photovoltaic power plants and the power grid will become more intelligent and automated. The integrated and optimized operation of distributed photovoltaic power plants will help improve the energy efficiency and reliability of the entire power grid. Therefore, the use of advanced communication technologies and data analysis methods to achieve remote monitoring and control of photovoltaic power plants and timely adjust active and reactive power output has become a key research area in smart grids. Summary of the Invention

[0004] The present invention combines the existing composition structure of the energy storage power station and uses a machine learning algorithm model to monitor the power of the power grid, predict changes in the power of the power grid, automatically control the operation of relays, connect to the energy storage battery, and use the charging and discharging of the energy storage battery to regulate the power of the power grid.

[0005] The technical solution provided by the present invention is: a method for remote monitoring, interconnection and control of photovoltaic energy storage power stations, the method comprising:

[0006] A method for remotely monitoring and controlling a photovoltaic energy storage power station, comprising the following steps:

[0007] The data acquisition device collects the status data of the energy storage battery, grid power data and relay status data, and sends them to the cloud through the Internet of Things;

[0008] Obtain battery status data and grid power data from the cloud, analyze the battery status and grid power data using a power monitoring model, predict the grid power status, and perform reactive power compensation and voltage regulation.

[0009] By integrating relay status data, battery status data, and grid power data into a fusion monitoring model, the relay switching status is predicted, and the normal connection or disconnection of the energy storage battery to the grid and / or photovoltaic power generation unit is automatically controlled;

[0010] The status data of the energy storage battery includes battery temperature, battery power, and battery voltage;

[0011] The grid power data includes grid voltage, grid current, grid reactive power, and grid active power;

[0012] The relay status data includes relay current and relay switching status.

[0013] Preferably, the collecting of the energy storage battery status data, grid power data and relay status data by a data acquisition device and sending the data to the cloud via the Internet of Things comprises the following steps:

[0014] Arrange voltage sensors, current sensors, and temperature sensors with IoT communication functions inside the energy storage battery pack of the energy storage power station;

[0015] The voltage sensor, current sensor and temperature sensor collect battery voltage, battery current and battery temperature data at the set collection frequency, send the data to the battery data collection device through the Internet of Things, and then send it to the cloud through the battery data collection device;

[0016] A battery database is established in the cloud to store battery voltage, battery power and battery temperature data.

[0017] Preferably, the collecting of the energy storage battery status data, grid power data and relay status data by the data acquisition device and sending the data to the cloud via the Internet of Things further comprises the following steps:

[0018] Arrange current transformers, voltage transformers, and power sensors with IoT communication capabilities at the grid end;

[0019] The current transformer, voltage transformer, and power sensor collect grid current, grid voltage, grid reactive power, and grid active power at the set collection frequency, and send the data to the grid data collection device through the Internet of Things, and then send it to the cloud through the grid data collection device;

[0020] A power grid database is established in the cloud to store grid current, grid voltage, grid reactive power and grid active power data.

[0021] Preferably, the collecting of the energy storage battery status data, grid power data and relay status data by the data acquisition device and sending the data to the cloud via the Internet of Things further comprises the following steps:

[0022] Current sensors with IoT functions are respectively provided at the input and output ends of the relay;

[0023] The current sensor collects the current at the input and output terminals of the relay at the set collection frequency, sends it to the relay data collection device through the Internet of Things, and then sends it to the cloud through the relay data collection device;

[0024] A relay status database is established in the cloud to store the current data of the relay input and output ends and the relay switching status data.

[0025] Preferably, the analysis of the battery energy storage status and grid power data by the power monitoring model includes the following steps:

[0026] Build a power monitoring model: Use LSTM neural network to build a power monitoring model to predict the future power status of the power grid;

[0027] Acquiring historical power data of the power grid from a power grid database, wherein the historical power data includes reactive power data and active power data;

[0028] After pre-processing the reactive power data and the active power data, normalization is performed to form a reactive power data set and an active power data set;

[0029] Use a sliding time window to collect data in the active power data set to form an active power time series, and select multiple data from the active power time series to form a training sequence 1 and a corresponding output sequence 1;

[0030] Use a sliding window to collect data in the reactive power data set to form a reactive power time series, and select multiple data from the reactive power time series to form a training sequence 2 and a corresponding output sequence 2;

[0031] Use a neural network with multiple LSTM layers and dense output layers to build active power monitoring models and reactive power monitoring models;

[0032] Inputting a training sequence 1 and a corresponding output sequence 1 of a certain length into an active power monitoring model for training;

[0033] Inputting a training sequence 2 of a certain length and a corresponding output sequence 2 into the reactive power monitoring model for training;

[0034] Input the latest active power data of a certain period of time into the trained power monitoring model, and output the predicted active power value for a certain period of time in the future. ;

[0035] Input the reactive power data of the latest period of time into the trained reactive power monitoring model, and output the reactive power prediction value for the next period of time ;

[0036] Setting the active power threshold and reactive power threshold ;

[0037] When the active power value is predicted , it is believed that the active power of the power grid will be normal in the future, They are active power convergence value and reactive power convergence value respectively;

[0038] if or It is considered that the power of the power grid will be abnormal at some future time, and needs to be adjusted, and the corresponding abnormal time is generated. Status value , used to identify the state of the grid power, the state value .

[0039] Preferably, the predicting of the power state of the power grid and performing reactive power compensation and voltage regulation comprises the following steps:

[0040] if , it is considered that the active power of the grid is too high, and the excess power is absorbed by the energy storage battery, including the following steps:

[0041] Connect relay 1 between the converter and the power station output bus;

[0042] Connect relay 2 between the combiner box and the converter;

[0043] when When the PV power generation unit is on, the first relay is disconnected and the second relay is closed, and the energy storage battery pack is charged through the photovoltaic power generation unit;

[0044] if , it is considered that the reactive power of the power grid is too high, and the energy storage battery is discharged through the converter to adjust the reactive power; including the following steps:

[0045] when When , relay 1 is closed and relay 2 is open;

[0046] The converter adjusts the phase of the output voltage to create a phase difference between the current and voltage, thereby absorbing or compensating for the reactive power of the power grid.

[0047] Preferably, the method of fusing relay status data, battery status data and grid power data by fusing a monitoring model to predict the switching state of the relay comprises the following steps:

[0048] Constructing a fusion monitoring model: Constructing a fusion monitoring model through regression algorithm;

[0049] Obtain active power, reactive power, and battery power at multiple times, and after normalization, form a fusion data matrix ,in Respectively indicate time Active power, reactive power and energy storage battery capacity, ;

[0050] Each row of the fusion matrix is ​​a fusion data vector. The fusion data vectors at different times are used as the input of the model and input into the fusion monitoring model. The output of the fusion monitoring model is Time relay Predicted value of switching status ,in ; , is the number of relays;

[0051] when When the corresponding relay is disconnected, , the corresponding relay is closed.

[0052] Preferably, the fusion monitoring model is constructed by a regression algorithm, and the fusion monitoring model is:

[0053] ,in is the intercept, are the regression coefficients, To adjust the parameters, For the moment Time relay The switching state prediction value of .

[0054] The automatic control of the normal connection or disconnection of the energy storage battery with the power grid and / or photovoltaic power generation unit comprises the following steps:

[0055] Building a relay control matrix , ;

[0056] Control the matrix through the relay to control the time The corresponding control relay is closed and opened to realize the connection and disconnection between the photovoltaic power generation unit and the energy storage battery. When the relay Disconnect, when When the relay closure.

[0057] The present invention also provides a photovoltaic energy storage power station remote monitoring, Internet of Things, and control system, including several voltage sensors, several current sensors, several temperature sensors, several current transformers, several voltage transformers, several power sensors, several relays 1 and 2, several power grid data acquisition devices, several battery data acquisition devices, and a cloud server. The current transformers, voltage transformers, and power sensors are respectively communicatively connected to the power grid data acquisition devices, the relays 1 and 2, voltage sensors, current sensors, and temperature sensors are respectively connected to the battery data acquisition devices, the power grid data acquisition devices and the battery data acquisition devices are respectively connected to the cloud server, and the system is used to execute the photovoltaic energy storage power station remote monitoring, Internet of Things, and control method.

[0058] The present invention also provides a computer-readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the photovoltaic energy storage power station remote monitoring, Internet of Things, and Control method.

[0059] Beneficial effects of the present invention:

[0060] 1. The present invention uses a data acquisition device to collect energy storage battery status data, grid power data, and relay status data, and sends it to the cloud via the Internet of Things. The energy storage battery status data and grid power data are obtained from the cloud, and the battery energy storage status and grid power data are monitored and analyzed using a power monitoring model to predict the grid power status. Combined with Internet of Things technology, the analysis and monitoring processes can be achieved remotely.

[0061] 2. The present invention uses LSTM neural network to build a power monitoring model to predict the future power state of the power grid. The active power data of the latest period of time is input into the trained power monitoring model, and the future active power prediction value is output; the reactive power data of the latest period of time is input into the trained reactive power monitoring model, and the reactive power prediction value of the future period of time is output; and the active power threshold is set. and reactive power threshold ;when When , relay 1 is closed, relay 2 is disconnected, and the photovoltaic power generation unit charges the energy storage battery through relay 2; if , it is considered that the reactive power of the power grid is too high, and the energy storage battery adjusts the reactive power by discharging through the converter; it realizes the function of automatic power adjustment when a power grid fault is predicted.

[0062] 3. The present invention constructs a fusion monitoring model, which is respectively: ; The output of the model is the switching status of relay 1 and relay 2 (open, closed). Through the trained fusion monitoring model, we established the relationship between active power, reactive power, battery power and the switching status of relay 1 and relay 2 during the fault. Based on these output values, a relay control matrix was constructed. When a power grid fault occurs, the matrix can automatically control relay 1 and relay 2 to operate, thereby automatically establishing a discharge circuit and a charging circuit, which ensures the effective implementation of the power regulation function.

[0063] 4. When constructing the fusion monitoring model, the present invention further considers the situation where the power of the energy storage battery is too low to be discharged; in this case, Transformed into ,in is the battery power coefficient; when the current battery power is less than the minimum power, =0, when the current battery power is greater than the minimum power, =1; when =0, Once the relay is disconnected, the energy storage battery cannot be discharged. The system will select other energy storage batteries with sufficient charge for discharge, giving priority to the energy storage battery pack with the highest battery capacity. This solves the problem of not being able to determine which energy storage battery pack is involved in the charging and discharging process, and avoids the problem of being unable to adjust power due to low-charge energy storage batteries being unable to meet discharge requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 Flow chart of the method of the present invention;

[0065] Figure 2 This is a flow chart of the power monitoring and regulation process of the present invention;

[0066] Figure 3 This is a schematic diagram of the connection of photovoltaic power station equipment of the present invention. DETAILED DESCRIPTION

[0067] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are for illustrative purposes only, and those skilled in the art will readily appreciate other obvious variations. The basic principles of the present invention defined in the following description may be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the present invention.

[0068] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the elements may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0069] Please combine Figure 1 The present invention provides a method for remote monitoring and control of photovoltaic energy storage power stations.

[0070] First, the data acquisition device collects the status data of the energy storage battery, grid power data, and relay status data, and sends it to the cloud through the Internet of Things;

[0071] Specifically:

[0072] S101. Deploy current transformers, voltage transformers, and power sensors with IoT communication capabilities at the power grid end. The current transformers, voltage transformers, and power sensors collect grid current, grid voltage, grid reactive power, and grid active power at a set collection frequency, transmit the data to a grid data collection device via the IoT, and then transmit the data to the cloud via the grid data collection device. Establish a grid database in the cloud to store grid current, grid voltage, grid reactive power, and grid active power data. The grid end in this embodiment refers to the output end of the photovoltaic energy storage power station (the output end of the step-up transformer). The collection frequency can be selected based on actual needs, for example, once every minute or once every hour. The grid data collection device, acting as a data transfer station, can perform preliminary data processing, such as format conversion and compression.

[0073] S102. Current sensors with IoT functionality are installed at the input and output ends of the relays. The relays here include relay 1 (S1) and relay 2 (S2) between the converter and the combiner box, and relay S3 between the photovoltaic inverter and the photovoltaic power station's grid output busbar. The current sensors are selected to have IoT communication capabilities. In this example, the current sensors are wirelessly connected to a grid data acquisition device. Data transmitted by the current sensors is transmitted to the grid data acquisition device, which then sends it to a cloud server, accommodating distributed, multi-type data collection and centralized transmission. Relay status data includes relay current (input and output current) and relay switching status (disconnected and connected).

[0074] The current sensor collects the current at the input and output ends of the relay at the set collection frequency, sends it to the relay data collection device through the Internet of Things, and then sends it to the cloud through the relay data collection device. By collecting the current at the input and output ends of the relay, it is judged whether the relay is normal. When the input and output current values ​​of the relay are 0, it means that the relay is normally disconnected. When the input and output current values ​​are the same, it means that the relay is normally closed.

[0075] S103. Dispose voltage sensors, current sensors, and temperature sensors with Internet of Things communication functions in the energy storage battery pack of the energy storage power station; the voltage sensors, current sensors, and temperature sensors collect battery voltage, battery current, and battery temperature data at a set collection frequency, and send the data to a battery data collection device via the Internet of Things, and then send the data to the cloud via the battery data collection device.

[0076] S104. Establish a relay status database in the cloud to store the current data of the relay input and output terminals for later use; establish a power grid database to store the power grid current, power grid voltage, power grid reactive power and power grid active power data.

[0077] Then, the energy storage battery status data and grid power data are obtained from the cloud. The battery energy storage status and grid power data are analyzed through the power monitoring model to predict the grid power status, perform reactive power compensation and voltage regulation.

[0078] Specifically:

[0079] S201. Use an LSTM neural network to build a power monitoring model to predict the future power state of the power grid; that is, use a neural network with multiple LSTM layers and dense output layers to build an active power monitoring model and a reactive power monitoring model.

[0080] Historical power data of the power grid is obtained from a power grid database, wherein the historical power data includes reactive power data and active power data; the reactive power data and active power data are preprocessed, including steps such as cleaning, removing outliers, filling missing values, and formatting for subsequent processing; and the preprocessed data are normalized to form a reactive power data set and an active power data set.

[0081] For example, if the data is sampled once a minute and the system wants to capture hourly trends, a 60-minute window is selected. Starting from the beginning of the data set, the selected window size is applied to the data sequence, and the window is slid element by element until the window reaches the end of the data set. At each step, the data within the window constitutes the current time series segment. Based on actual needs, the time step used for future predictions is determined. For example, if you want to predict the active power change in the next 10 minutes, the output sequence is the active power value 10 minutes after each time window. In each time window, the data within the window is used as the training sequence, and the active power data within the subsequent predetermined time step is used as the output sequence. A large number of data pairs are obtained: (training sequence one, output sequence one), which constitute the training data set.

[0082] In the active power monitoring model, LSTM layers are stacked, each with a different number of neurons. After the LSTM layer, one or more dense (fully connected) layers are added to process the output of the LSTM layer and generate the final prediction result.

[0083] For example, in this embodiment, our goal is to predict the active power value in the next 10 minutes. The output layer is designed as a dense layer with a single neuron to produce a single continuous value.

[0084] Then, an Adam optimizer is selected for the active power monitoring model, and the mean square error (MSE) is selected as the loss function.

[0085] The training sequence 1 and the output sequence 1 are input into the active power monitoring model. The active power monitoring model is trained using the training data, and the network weights are adjusted through multiple iterative learning.

[0086] The construction and training process of the reactive power monitoring model can refer to the above process.

[0087] S202: Input the latest active power data of a certain period of time into the trained power monitoring model, and output the predicted active power value for a certain period of time in the future. ;

[0088] Input the reactive power data of the latest period of time into the trained reactive power monitoring model, and output the reactive power prediction value for the next period of time ;

[0089] Setting the active power threshold and reactive power threshold ;

[0090] When the active power value is predicted , it is believed that the active power of the power grid will be normal in the future, They are active power convergence value and reactive power convergence value respectively;

[0091] if or It is considered that the power of the power grid will be abnormal at some future time, and needs to be adjusted, and the corresponding abnormal time is generated. Status value , used to identify the state of the grid power, the state value .when Shi, indicating the time The power grid is abnormal. On the contrary, it means that the power grid is abnormal at the time normal.

[0092] S203, performing reactive power compensation and voltage regulation, including the following steps:

[0093] S2031, if , the active power output of the energy storage power station grid is too high, and the excess power is absorbed by the energy storage battery, including the following steps:

[0094] Connect relay 1 (S1) between the grid and the converter;

[0095] Connect relay 2 (S2) between the combiner box and the converter;

[0096] when When , relay 1 (S1) is disconnected and relay 2 (S2) is closed. The photovoltaic power generation unit charges the energy storage battery through relay 2, reducing the current input to the power station output bus;

[0097] S2032, if , it is considered that the reactive power of the power grid is too high, and the energy storage battery is discharged through the converter to adjust the reactive power; including the following steps:

[0098] when When , relay 1 (S1) is closed and relay 2 (S2) is open;

[0099] The converter adjusts the phase of the output voltage to create a phase difference between the current and voltage, thereby absorbing or compensating for the reactive power of the power grid.

[0100] The switching actions of the relays S1 and S2 are realized by integrating the output values ​​of the monitoring model.

[0101] Specifically, construct the relay control matrix , ;

[0102] Each element value in the relay control matrix is ​​the output value of the fusion monitoring model, which is: ;

[0103] in is the intercept, are the regression coefficients, To adjust the parameters, For the moment Time relay The switching state prediction value of .

[0104] In this embodiment, the fusion monitoring models of relay 1 (S1) and relay 2 (S2) that need to be controlled are:

[0105] ; ,

[0106] Considering the situation where the power of the energy storage battery is too low to be discharged; at this time,

[0107] Will Transformed into ,in is the battery capacity factor.

[0108] The active power, reactive power, battery power and relay switching status data when the power grid fails are obtained from the database. Active power, reactive power and battery power are used as input variables, and relay switching status is used as output set. Active power, reactive power and battery power constitute a fusion data matrix. ,in Respectively indicate time Active power, reactive power and energy storage battery capacity, The model associates active power, reactive power, and battery charge with the relay status, comprehensively considering multiple data points, including the grid's active power, reactive power, and battery charge. This model incorporates battery charge to avoid situations where a battery is insufficient and unable to effectively complete a discharge task. PV energy storage power stations typically have multiple PV energy storage systems, each of which includes at least one PV power generation unit and a corresponding energy storage unit (energy storage battery pack). When the charge of one energy storage battery pack is insufficient to meet discharge requirements, discharge can be carried out using other battery packs with sufficient charge.

[0109] In this embodiment, the minimum power of the energy storage battery is set. When the power of the current battery is less than the minimum power, =0, when the current battery power is greater than the minimum power, =1;

[0110] when =0, , relay 1 (S1) is disconnected, and the energy storage battery is in a state where it cannot be discharged. Then other energy storage batteries with sufficient power are selected for discharge. The principle of energy storage battery selection is: give priority to the energy storage battery pack with the highest battery.

[0111] The historical active power, reactive power, and battery charge of relay 1 (S1) and relay 2 (S2) obtained from the database are used as input variables, and the switching status of the two relays is obtained as the output set. The fusion monitoring model is trained using the existing regression model training method, which will not be repeated here.

[0112] Here we explain the principle of selecting the output value of the fusion monitoring model. When ,when , then it is believed that .

[0113] In this embodiment, the two relays that need to be controlled are relay S1 and relay S2, so the relay control matrix is ;

[0114] For example, express:

[0115] At time 2, , relay S1 is disconnected and relay S2 is closed, then the charging circuit is connected and the energy storage battery pack is charged through the power station output bus;

[0116] At time 3, , relay S1 is closed and relay S2 is disconnected. At this time, the discharge circuit is connected and the energy storage battery pack is discharged.

[0117] When active power or reactive power is abnormal, two fusion monitoring models are used to control the actions of relay one and relay two respectively, establish the charging circuit and discharging circuit of the energy storage battery, and use the charging and discharging of the energy storage battery to regulate the power of the grid.

[0118] Each row of the fusion matrix is ​​a fusion data vector. The fusion data vectors at different times are used as the input of the model and input into the fusion monitoring model. The output of the fusion monitoring model is Time relay Predicted value of switching status ,in ; , is the number of relays.

[0119] The present invention also provides a photovoltaic energy storage power station remote monitoring, Internet of Things, and physical control system, including several voltage sensors, several current sensors, several temperature sensors, several current transformers, several voltage transformers, several power sensors, several relays 1, relay 2, several power grid data acquisition devices, several battery data acquisition devices, and a cloud server. The current transformers, voltage transformers, and power sensors are respectively communicatively connected to the power grid data acquisition devices, the relays 1, relay 2, voltage sensors, current sensors, and temperature sensors are respectively connected to the battery data acquisition devices, the power grid data acquisition devices and the battery data acquisition devices are respectively connected to the cloud server, and the system is used to execute a photovoltaic energy storage power station remote monitoring, Internet of Things, and physical control method.

[0120] The present invention also provides a computer-readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the photovoltaic energy storage power station remote monitoring, Internet of Things, and Control method.

[0121] The embodiments disclosed in the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. The embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by the central processing unit (CPU), the above-mentioned functions defined in the method of the present application are executed. It should be noted that the computer-readable medium mentioned above in the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, a system, device or component of an electrical, magnetic, optical, electromagnetic, infrared segment, or semiconductor, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more wire segments, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including, but not limited to, an electromagnetic signal, an optical signal, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to wireless, electrical wire, optical fiber cable, RF, etc., or any suitable combination thereof.

[0122] The flow charts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0123] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments. Without departing from the principles, the implementation methods of the present invention may be subject to any deformation or modification.

Claims

1. A method for remote monitoring and control of photovoltaic energy storage power stations, characterized in that: The method comprises the following steps: The data acquisition device collects the status data of the energy storage battery, grid power data and relay status data, and sends them to the cloud through the Internet of Things; Obtain battery status data and grid power data from the cloud, and analyze the battery status and grid power data using a power monitoring model, including: Build a power monitoring model: Use LSTM neural network to build a power monitoring model to predict the future power status of the power grid; Acquiring historical power data of the power grid from a power grid database, wherein the historical power data includes reactive power data and active power data; After pre-processing the reactive power data and the active power data, normalization is performed to form a reactive power data set and an active power data set; Use a sliding time window to collect data in the active power data set to form an active power time series, and select multiple data from the active power time series to form a training sequence 1 and a corresponding output sequence 1; Use a sliding window to collect data in the reactive power data set to form a reactive power time series, and select multiple data from the reactive power time series to form a training sequence 2 and a corresponding output sequence 2; Use a neural network with multiple LSTM layers and dense output layers to build active power monitoring models and reactive power monitoring models; Inputting a training sequence 1 and a corresponding output sequence 1 of a certain length into an active power monitoring model for training; Inputting a training sequence 2 of a certain length and a corresponding output sequence 2 into the reactive power monitoring model for training; Input the latest active power data of a certain period of time into the trained power monitoring model, and output the predicted active power value for a certain period of time in the future. ; Input the reactive power data of the latest period of time into the trained reactive power monitoring model, and output the reactive power prediction value for the next period of time ; Setting the active power threshold and reactive power threshold ; When the active power value is predicted , it is believed that the active power of the power grid will be normal in the future, They are active power convergence value and reactive power convergence value respectively; if or It is considered that the power of the power grid will be abnormal at some future time, and needs to be adjusted, and the corresponding abnormal time is generated. Status value , used to identify the state of the grid power, the state value ; if , it is considered that the active power of the grid is too high, and the excess power is absorbed by the energy storage battery, including the following steps: Connect relay 1 between the converter and the power station output bus; Connect relay 2 between the combiner box and the converter; when When the PV power generation unit is on, the first relay is disconnected and the second relay is closed, and the energy storage battery pack is charged through the photovoltaic power generation unit; if , it is considered that the reactive power of the power grid is too high, and the energy storage battery is discharged through the converter to adjust the reactive power; including the following steps: when When , relay 1 is closed and relay 2 is open; The converter adjusts the phase of the output voltage to create a phase difference between the current and voltage, thereby absorbing or compensating for the reactive power of the grid. Constructing a fusion monitoring model: Constructing a fusion monitoring model through regression algorithm; Obtain active power, reactive power, and battery power at multiple times, and after normalization, form a fusion data matrix ,in Respectively indicate time Active power, reactive power and energy storage battery capacity, ; Each row of the fusion matrix is ​​a fusion data vector. The fusion data vectors at different times are used as the input of the model and input into the fusion monitoring model. The fusion monitoring model is: ,in is the intercept, are the regression coefficients, To adjust the parameters, For the moment Time relay The switching state prediction value of By integrating relay status data, battery status data, and grid power data into a fusion monitoring model, the relay switching status is predicted, and the normal connection or disconnection of the energy storage battery to the grid and / or photovoltaic power generation unit is automatically controlled; The automatic control of the normal connection or disconnection of the energy storage battery with the power grid and / or photovoltaic power generation unit comprises the following steps: Building a relay control matrix , ; Control the matrix through the relay to control the time The corresponding control relay is closed and opened to realize the connection and disconnection between the photovoltaic power generation unit and the energy storage battery. When the relay Disconnect, when When the relay closure; The output of the fusion monitoring model is Time relay Predicted value of switching status ,in ; , is the number of relays; when When the corresponding relay is disconnected, When , the corresponding relay is closed; The status data of the energy storage battery includes battery temperature, battery power, and battery voltage; The grid power data includes grid voltage, grid current, grid reactive power, and grid active power; The relay status data includes relay current and relay switching status.

2. The photovoltaic energy storage power station remote monitoring and IoT control method according to claim 1, characterized in that: The method of collecting the energy storage battery status data, grid power data and relay status data by the data acquisition device and sending them to the cloud via the Internet of Things includes the following steps: Arrange voltage sensors, current sensors, and temperature sensors with IoT communication functions inside the energy storage battery pack of the energy storage power station; The voltage sensor, current sensor and temperature sensor collect battery voltage, battery current and battery temperature data at the set collection frequency, send the data to the battery data collection device through the Internet of Things, and then send it to the cloud through the battery data collection device; A battery database is established in the cloud to store battery voltage, battery power and battery temperature data.

3. The photovoltaic energy storage power station remote monitoring and IoT control method according to claim 1, characterized in that: The data acquisition device is used to collect the status data of the energy storage battery, the power data of the power grid, and the status data of the relay, and the data is sent to the cloud through the Internet of Things, which further includes the following steps: Arrange current transformers, voltage transformers, and power sensors with IoT communication capabilities at the grid end; The current transformer, voltage transformer, and power sensor collect grid current, grid voltage, grid reactive power, and grid active power at the set collection frequency, and send the data to the grid data collection device through the Internet of Things, and then send it to the cloud through the grid data collection device; A power grid database is established in the cloud to store grid current, grid voltage, grid reactive power and grid active power data.

4. The photovoltaic energy storage power station remote monitoring and IoT control method according to claim 1, characterized in that: The data acquisition device is used to collect the status data of the energy storage battery, the power data of the power grid, and the status data of the relay, and the data is sent to the cloud through the Internet of Things, which further includes the following steps: Current sensors with IoT functions are respectively provided at the input and output ends of the relay; The current sensor collects the current at the input and output terminals of the relay at the set collection frequency, sends it to the relay data collection device through the Internet of Things, and then sends it to the cloud through the relay data collection device; A relay status database is established in the cloud to store the current data of the relay input and output ends and the relay switching status data.

5. A photovoltaic energy storage power station remote monitoring, Internet of Things, and control system, comprising a plurality of voltage sensors, a plurality of current sensors, a plurality of temperature sensors, a plurality of current transformers, a plurality of voltage transformers, a plurality of power sensors, a plurality of relays 1 and 2, a plurality of power grid data acquisition devices, a plurality of battery data acquisition devices, and a cloud server. The current transformers, voltage transformers, and power sensors are respectively connected to the power grid data acquisition devices in a communication manner. The relays 1 and 2, voltage sensors, current sensors, and temperature sensors are respectively connected to the battery data acquisition devices. The power grid data acquisition devices and battery data acquisition devices are respectively connected to the cloud server. The system is used to execute the photovoltaic energy storage power station remote monitoring, Internet of Things and Control method described in any one of claims 1 to 4 above.

6. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the photovoltaic energy storage power station remote monitoring, Internet of Things, and Control method according to any one of claims 1 to 4.

Images