Monitoring method and device of energy storage equipment, electronic equipment and storage medium

Abstract

The invention discloses a monitoring method and device of energy storage equipment, electronic equipment and a storage medium. The method comprises: determining a distributed device, the distributed device comprising a plurality of first energy storage devices; and for each first energy storage device, determining real-time energy storage time sequence data of the first energy storage device, predicting the input real-time energy storage time sequence data through a first prediction model to obtain predicted energy storage time sequence data, and determining a fault detection result of the first energy storage device according to the predicted energy storage time sequence data, the first prediction model is obtained by training a neural network through a first training sample; and determining a monitoring result of the distributed equipment according to the plurality of fault detection results. The effect of performing high-accuracy centralized fault detection on the multiple distributed energy storage devices is achieved, and high-quality unified monitoring of the multiple distributed energy storage devices by the cloud monitoring end is achieved.

Description

Technical Field

[0001] The present invention relates to the field of computer application technologies, and particularly to a monitoring method, device, electronic device, and storage medium for energy storage devices. Background Art

[0002] With the rapid development of renewable energy, energy storage systems (such as battery energy storage or supercapacitors, etc.) are widely used in scenarios such as power grid frequency modulation, peak shaving and valley filling, and emergency power supply. Therefore, to meet social needs, energy storage devices are also distributed and constructed at different locations.

[0003] Currently, a local system for fault monitoring of energy storage devices is set up at each location. Through the local system, the operation data of local energy storage devices can be collected, and abnormal analysis can be performed by comparing with thresholds to carry out device fault alarms. However, the data abnormal analysis method is relatively single, so it is difficult to ensure the accuracy of device fault detection, and there is a data island problem among multiple local systems, and unified fault monitoring of multiple distributed energy storage devices cannot be achieved. That is, currently, there is a lack of a centralized fault monitoring method with high accuracy for multiple distributed energy storage devices. Summary of the Invention

[0004] The present invention provides a monitoring method, device, electronic device, and storage medium for energy storage devices to solve the current lack of a centralized fault monitoring method with high accuracy for multiple distributed energy storage devices.

[0005] According to one aspect of the present invention, a monitoring method for energy storage devices is provided. The method includes:

[0006] Determine distributed devices, where the distributed devices include multiple first energy storage devices;

[0007] For each of the first energy storage devices, determine the real-time energy storage time series data of the first energy storage device, predict the input real-time energy storage time series data through a first prediction model to obtain predicted energy storage time series data, and determine the fault detection result of the first energy storage device according to the predicted energy storage time series data, where the first prediction model is obtained by training a neural network with a first training sample;

[0008] Determine the monitoring result of the distributed devices according to the multiple fault detection results.

[0009] According to another aspect of the present invention, a monitoring device for energy storage devices is provided. The device includes:

[0010] A distributed device determination module, configured to determine distributed devices, where the distributed devices include multiple first energy storage devices;

[0011] The device failure detection module is used to determine the real-time energy storage timing data of each of the first energy storage devices, predict the input real-time energy storage timing data through a first prediction model to obtain predicted energy storage timing data, and determine the failure detection result of the first energy storage device according to the predicted energy storage timing data, where the first prediction model is obtained by training a neural network with first training samples;

[0012] The monitoring result determination module is used to determine the monitoring result of the distributed device according to multiple failure detection results.

[0013] According to another aspect of the present invention, there is provided an electronic device, which includes:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the monitoring method of the energy storage device according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the monitoring method of the energy storage device according to any embodiment of the present invention when executed.

[0018] The technical solution of the embodiment of the present invention determines a distributed device, where the distributed device includes a plurality of first energy storage devices; for each of the first energy storage devices, determines the real-time energy storage timing data of the first energy storage device, predicts the input real-time energy storage timing data through a first prediction model to obtain predicted energy storage timing data, and determines the failure detection result of the first energy storage device according to the predicted energy storage timing data, where the first prediction model is obtained by training a neural network with first training samples; determines the monitoring result of the distributed device according to multiple failure detection results. The present invention achieves the effect of performing high-accuracy centralized failure detection on multiple distributed energy storage devices, and realizes high-quality unified monitoring of multiple distributed energy storage devices by the total monitoring end.

[0019] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. Description of the Drawings

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0021] Figure 1 is a flowchart of a monitoring method for an energy storage device provided in Embodiment 1 of the present invention;

[0022] Figure 2 is a flowchart of a monitoring method for an energy storage device provided in Embodiment 2 of the present invention;

[0023] Figure 3 is a flowchart of a monitoring method for an energy storage device provided in Embodiment 3 of the present invention;

[0024] Figure 4 is a schematic diagram of the architecture of a monitoring system for an energy storage device provided in the embodiments of the present invention;

[0025] Figure 5 is an overall flowchart of a monitoring method for an energy storage device provided in the embodiments of the present invention;

[0026] Figure 6 is a schematic structural diagram of a monitoring device for an energy storage device provided in Embodiment 4 of the present invention;

[0027] Figure 7 is a schematic structural diagram of an electronic device for implementing the monitoring method of the energy storage device in the embodiments of the present invention. Detailed implementation manners

[0028] To enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, rather than all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0029] It should be noted that the terms "first", "second", etc. in the description, claims and above-mentioned drawings of the present invention are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0030] Embodiment 1

[0031] Figure 1 The figure is a flowchart of a monitoring method for an energy storage device provided in Embodiment 1 of the present invention. This embodiment is applicable to the situation of centralized monitoring of multiple distributed energy storage devices. This method can be executed by a monitoring device of the energy storage device. The monitoring device of the energy storage device can be implemented in the form of hardware and / or software, and the monitoring device of the energy storage device can be configured in a computer. As Figure 1 shown, the method includes:

[0032] S110. Determine distributed devices.

[0033] In the embodiment of the present invention, the distributed devices include multiple first energy storage devices.

[0034] Among them, the distributed devices can be understood as multiple energy storage devices distributed dispersedly. The first energy storage device can be understood as an energy storage device to be monitored for faults. Generally speaking, the distributed devices can be multiple energy storage devices located at different locations. The energy storage device can be understood as a device with energy storage function. Optionally, the energy storage device can be a battery or a supercapacitor and other related devices.

[0035] S120. For each of the first energy storage devices, determine the real-time energy storage time series data of the first energy storage device, predict the input real-time energy storage time series data through a first prediction model to obtain predicted energy storage time series data, and determine the fault detection result of the first energy storage device according to the predicted energy storage time series data.

[0036] In the embodiment of the present invention, the first prediction model is obtained by training a neural network with a first training sample.

[0037] Among them, the real-time energy storage time series data can be understood as the real-time time series of energy storage data. The real-time energy storage time series data may include real-time energy storage data and reference energy storage time series data. Generally speaking, the reference energy storage time series data can be understood as historical energy storage time series data.

[0038] Optionally, determining the real-time energy storage time series data of the first energy storage device includes:

[0039] Obtain the real-time energy storage data of the first energy storage device, where the energy storage data includes at least one of current, voltage, charging power, discharging power, temperature, state of charge, and health state;

[0040] Obtain the reference energy storage time series data corresponding to the first energy storage device, and determine the real-time energy storage time series data according to the reference energy storage time series data and the real-time energy storage data.

[0041] Among them, the real-time energy storage data can be understood as real-time energy storage data.

[0042] The real-time energy storage time series data may include multiple time series points and the energy storage data corresponding to each time series point. This time series point can be understood as a time point. The time interval between multiple time series points can be preset according to the scenario requirements and will not be specifically limited here. Exemplarily, this time interval can be 5s, 10s, or 20s, etc. If the time interval is 5s, the time period for obtaining the energy storage data is 5s. Specifically, obtain the real-time energy storage data of the energy storage device every 5s, and update the historically determined energy storage data sequence according to the real-time energy storage data to obtain the real-time energy storage time series data.

[0043] The first prediction model can be understood as a learning model with data prediction function. The first prediction model is obtained by training a neural network with a first training sample. The neural network can be understood as a computational model that simulates the human brain neuron network. In the embodiments of the present invention, the neural network can be preset according to the scenario requirements and will not be specifically limited here. Optionally, the neural network can be a long short-term memory network. The long short-term memory network (Long Short-Term Memory, LSTM) can be understood as a recurrent neural network that can solve the problem of gradient disappearance faced when processing long sequences. Specifically, input the real-time energy storage time series data into the first prediction model to obtain the energy storage data of the predicted time series point output by the first prediction model. Generally speaking, the predicted time series point can be understood as the adjacent future time series point corresponding to the real-time time series point for obtaining the real-time energy storage data. Exemplarily, if the real-time time series point is time point a, the predicted time series point can be time point b corresponding to a + 5s.

[0044] The predicted energy storage time series data can be understood as the predicted energy storage data sequence. The predicted energy storage time series data may include the predicted energy storage data, the real-time energy storage data, and the reference energy storage sequence data.

[0045] The fault detection result can characterize whether there is a fault in the first energy storage device. Optionally, the fault detection result may include having a fault or not having a fault. In the embodiments of the present invention, the fault detection results corresponding to different first energy storage devices may be the same or different.

[0046] S130. Determine the monitoring result of the distributed device according to multiple fault detection results.

[0047] Among them, the monitoring result can be understood as the result of centralized monitoring of the distributed device.

[0048] Optionally, for the monitoring method of the energy storage device, after determining the monitoring result of the distributed device according to multiple fault detection results, it further includes:

[0049] Display the monitoring result through a target display screen.

[0050] Based on the above embodiment solution, visual centralized monitoring of distributed energy storage devices is achieved.

[0051] The technical solution of the embodiments of the present invention determines a distributed device, where the distributed device includes multiple first energy storage devices; for each first energy storage device, determines the real-time energy storage time series data of the first energy storage device, predicts the input real-time energy storage time series data through a first prediction model to obtain predicted energy storage time series data, and determines the fault detection result of the first energy storage device according to the predicted energy storage time series data, where the first prediction model is obtained by training a neural network with a first training sample; determines the monitoring result of the distributed device according to multiple fault detection results. The present invention achieves the effect of high-accuracy centralized fault detection for multiple distributed energy storage devices, and realizes high-quality unified monitoring of multiple distributed energy storage devices by the cloud monitoring end.

[0052] Embodiment Two

[0053] Figure 2 This is a flowchart of a monitoring method for an energy storage device provided in the second embodiment of the present invention. This embodiment refines the step of predicting the input real-time energy storage time series data through the first prediction model to obtain predicted energy storage time series data in the above embodiment. As Figure 2 shown, the method includes:

[0054] S210. Determine the distributed device.

[0055] S220. For each of the first energy storage devices, determine the real-time energy storage timing data of the first energy storage device.

[0056] S230. Perform energy storage data prediction on the input real-time energy storage timing data through the first prediction model to obtain predicted energy storage data.

[0057] Wherein, the predicted energy storage data can be understood as the energy storage data of the predicted timing points obtained by prediction.

[0058] S240. Determine the predicted energy storage timing data according to the predicted energy storage data and the real-time energy storage timing data.

[0059] In the embodiment of the present invention, the predicted energy storage timing data includes the predicted energy storage data and the real-time energy storage timing data.

[0060] S250. Determine the fault detection result of the first energy storage device according to the predicted energy storage timing data.

[0061] Optionally, the predicted energy storage timing data includes the predicted energy storage data, real-time energy storage data, and reference energy storage sequence data; the determining the fault detection result of the first energy storage device according to the predicted energy storage timing data includes:

[0062] Perform anomaly detection on the real-time energy storage data to obtain a first detection result; and perform anomaly detection on the predicted energy storage data to obtain a second detection result;

[0063] Determine a third detection result of the reference energy storage data adjacent to the real-time energy storage data in time sequence in the reference energy storage sequence data;

[0064] Determine the fault detection result of the first energy storage device according to the first detection result, the second detection result, and the third detection result.

[0065] Wherein, the first detection result can characterize whether the real-time energy storage data is abnormal. The second detection result can characterize whether the predicted energy storage data is abnormal. The third detection result can characterize whether the reference energy storage data is abnormal.

[0066] The reference energy storage data can be understood as the energy storage data of the reference timing points. Generally speaking, the reference timing points can be understood as historical timing points. Exemplarily, the real-time timing point is the a time point, the predicted timing point can be the b time point corresponding to a + 5s, and the reference timing point can be the c time point corresponding to a - 5s.

[0067] Optionally, the first detection result may include normal data or abnormal data. The second detection result may include normal data or abnormal data. The third detection result may include normal data or abnormal data.

[0068] Specifically, determining the fault detection result of the first energy storage device according to the first detection result, the second detection result, and the third detection result may include:

[0069] When the first detection result is normal data, the second detection result is normal data, and the third detection result is normal data, determining the fault detection result of the first energy storage device as having a fault.

[0070] That is, if three consecutive abnormal energy storage data appear in the predicted energy storage time series data, it is considered that the first energy storage device has a fault.

[0071] In the embodiments of the present invention, the specific method for detecting anomalies in energy storage data is not specifically limited herein and can be preset according to scenario requirements. In summary, the energy storage data may include at least one of current, voltage, charging power, discharging power, temperature, state of charge, and health state. Specifically, the specific method for detecting anomalies in energy storage data is to detect anomalies in energy storage data based on pre-set relevant thresholds, etc.

[0072] Based on the above embodiment solution, by first predicting energy storage data and then determining whether there are three consecutive abnormal energy storage data in the predicted energy storage time series data to determine whether the first energy storage device is faulty, the timeliness and accuracy of fault detection of the energy storage device can be improved.

[0073] S260. Determine the monitoring result of the distributed device according to multiple fault detection results.

[0074] The technical solution of the embodiments of the present invention predicts energy storage data for the input real-time energy storage time series data through the first prediction model to obtain predicted energy storage data; determines the predicted energy storage time series data according to the predicted energy storage data and the real-time energy storage time series data. The accuracy of energy storage data prediction is achieved.

[0075] Embodiment III

[0076] Figure 3 It is a flowchart of a monitoring method for an energy storage device provided in Embodiment III of the present invention. This embodiment is an addition to the determination of the fault detection result of the first energy storage device according to the predicted energy storage time series data in the above embodiment. As Figure 3 shown, the method includes:

[0077] S310. Determine the distributed devices.

[0078] S320. For each of the first energy storage devices, determine the real-time energy storage timing data of the first energy storage device, predict the input real-time energy storage timing data through a first prediction model to obtain predicted energy storage timing data, and determine a fault detection result of the first energy storage device according to the predicted energy storage timing data.

[0079] S330. When the fault detection result indicates the existence of a fault, determine energy storage abnormal data in the predicted energy storage timing data, and classify the input energy storage abnormal data through a first classification model to obtain a first fault category of the first energy storage device.

[0080] Wherein, the energy storage abnormal data can be understood as abnormal energy storage data. Optionally, determining the energy storage abnormal data in the predicted energy storage timing data includes: taking the reference energy storage data, the real-time energy storage data, and the predicted energy storage data as the energy storage abnormal data. That is, taking three consecutive abnormal energy storage data as the energy storage abnormal data.

[0081] The first classification model can be understood as a learning model with data classification function. The first classification model is obtained by training a random forest model with a second training sample. The random forest model can be understood as a learning model that combines multiple decision trees for classification.

[0082] The fault category can be understood as the category of device faults. Optionally, the fault category includes internal short circuit, thermal runaway, capacity attenuation, etc.

[0083] Based on the above embodiment solutions, it is possible to accurately classify the faults of faulty energy storage devices when it is determined that there are faults in the energy storage devices.

[0084] Optionally, after obtaining the first fault category of the first energy storage device, it further includes:

[0085] Determine maintenance suggestion data corresponding to the faulty first energy storage device through a pre-constructed category-maintenance mapping relationship library and the first fault category, wherein the category-maintenance mapping relationship library includes multiple candidate fault categories, multiple candidate maintenance suggestions, and the mapping relationship between the candidate fault categories and the candidate maintenance suggestions;

[0086] The determining the monitoring result of the distributed device according to the multiple fault detection results includes:

[0087] Determine the monitoring result of the distributed device according to the multiple fault detection results, the first fault category, and the maintenance suggestion data.

[0088] Among them, the category maintenance mapping relationship library can be understood as a database containing the mapping relationship between fault categories and maintenance suggestions.

[0089] The maintenance suggestion data can be understood as data related to fault maintenance suggestions. Optionally, the maintenance suggestion data may include checking the cooling system corresponding to the fault category of thermal runaway, etc.

[0090] In the implementation of the present invention, the candidate fault categories, the candidate maintenance suggestions, and the mapping relationship between the candidate fault categories and the candidate maintenance suggestions in the category maintenance mapping relationship library may be set in advance according to user requirements or scenario requirements, and no specific limitation is made here. The mapping relationship between the candidate fault categories and the candidate maintenance suggestions may be one-to-one, one-to-many, many-to-one, or many-to-many.

[0091] The candidate fault categories can be understood as candidate categories to be matched with the first fault category. Specifically, determine the target fault category that matches the first fault category among the multiple candidate fault categories, and use the candidate maintenance suggestions that have a mapping relationship with the target fault category as the maintenance suggestions corresponding to the first fault category.

[0092] Based on the above embodiment solution, after centralized fault detection and fault classification of distributed devices, automatic retrieval of maintenance methods is realized, without the need for monitors to manually retrieve maintenance methods after equipment failures occur. The present invention realizes fully automated centralized monitoring of distributed devices and timely feedback of maintenance suggestions after monitoring, effectively reducing the workload of monitors.

[0093] Optionally, the monitoring method of the energy storage device further includes: after determining the maintenance suggestion data corresponding to the first energy storage device with a fault through the pre-constructed category maintenance mapping relationship library and the first fault category, it further includes: using the fault detection result, the fault category, and the maintenance suggestion data corresponding to the first energy storage device as target data, and sending the target data to the distributed system corresponding to the first energy storage device.

[0094] S340. Determine the monitoring result of the distributed device according to the multiple fault detection results and the first fault category.

[0095] In the technical solution of the embodiment of the present invention, when the fault detection result indicates the existence of a fault, abnormal energy storage data in the predicted energy storage time series data is determined, and the input abnormal energy storage data is classified by a first classification model to obtain a first fault category of the first energy storage device, where the first classification model is obtained by training a random forest model with a second training sample; determining the monitoring result of the distributed device according to multiple fault detection results includes: determining the monitoring result of the distributed device according to multiple fault detection results and the first fault category. The present invention achieves the effect of accurately classifying faults of a faulty energy storage device when it has been determined that there is a fault in the energy storage device.

[0096] Figure 4 It is a schematic diagram of the architecture of a monitoring system for an energy storage device provided by an embodiment of the present invention. The following combines Figure 4 to introduce the monitoring system for an energy storage device to which the monitoring method for an energy storage device is applied. The architecture of this monitoring system may include a data acquisition layer, a data transmission layer, a cloud platform layer, and an application layer.

[0097] Among them, the data acquisition layer can collect real-time energy storage data through sensors installed on the energy storage device.

[0098] The data transmission layer can encrypt and transmit the energy storage data to the cloud through the 5th Generation Mobile Communication Technology (5G) or Narrowband Internet of Things technology.

[0099] The cloud platform layer may include a big data storage module, a fault diagnosis module, a decision-making module, and a visualization interface. Among them, the big data storage module can store a large amount of device energy storage data using a time series database. The fault diagnosis module can perform anomaly detection and fault classification on the energy storage data based on an LSTM neural network and a random forest model respectively. The visualization interface can display a device status dashboard, alarm push, and optimization suggestions through a target display device.

[0100] The application layer can support users to remotely control the device, download analysis reports, and issue instructions.

[0101] Furthermore, through the cloud platform layer, real-time charge and discharge strategies can be generated to balance energy storage efficiency, device life, and operating costs.

[0102] Specifically, a multi-objective optimization model is established in advance, and the objective function can include functions such as maximizing energy storage efficiency, minimizing life loss, and minimizing operating costs. The real-time adjustment of the charge and discharge strategy is carried out through the constructed objective function.

[0103] Through the cloud platform layer, the non-dominated sorting genetic algorithm can also be used to solve the optimal charging and discharging strategy. The non-dominated sorting genetic algorithm is further elaborated as follows: Algorithm parameters: population size = 100, crossover probability = 0.8, mutation probability = 0.1, maximum iteration = 200. Pareto front screening: Select the top ten non-dominated solutions using crowding distance sorting. Edge computing deployment: The policy generation is containerized (Docker image ≤ 50MB) and sent to the local programmable logic controller (PLC) through the Message Queuing Telemetry Transport (MQTT) protocol. Real-time verification: Ensure that the latency from data input to policy output is <500ms.

[0104] Furthermore, for the monitoring system of energy storage devices, remote operation and maintenance can be carried out. It supports updating the device firmware based on the Over-The-Air (OTA) technology and automatically dispatching maintenance tasks to the operation and maintenance personnel according to the diagnostic results, realizing the intelligent management of the entire life cycle of the device.

[0105] OTA firmware update. Differential update: Use the bsdiff algorithm to generate incremental packages (reducing the data volume transmitted by 90%). Security mechanism: Firmware signature: Verify the integrity of the update package through the elliptic curve digital signature algorithm. Rollback strategy: If the device heartbeat is lost after the update, automatically roll back to the previous stable version.

[0106] Figure 5 This is the overall flowchart of a monitoring method for energy storage devices provided by an embodiment of the present invention. The following further elaborates Figure 5 the overall process of the monitoring method for energy storage devices.

[0107] 1. Collect real-time energy storage data, perform preprocessing such as cleaning or standardization on the real-time energy storage data to obtain preprocessed real-time energy storage data, and determine real-time energy storage time series data based on the real-time energy storage data and historical energy storage data.

[0108] 2. Perform anomaly detection on the real-time energy storage time series data through the LSTM model. The LSTM model is further elaborated as follows:

[0109] Model architecture. Input layer: Time step = 60, feature dimension = 8 (the feature dimension can include dimensions such as voltage, current, and temperature). Hidden layer: The hidden layer includes 128 neurons / layer, Dropout = 0.2. Output layer: Predict the anomaly probability through the Sigmoid activation function.

[0110] Training strategy. Loss function: FocalLoss(γ = 2) is used to solve the problem of class imbalance. Optimizer: AdamW (learning rate = 1e-4, weight decay = 1e-5).

[0111] Inference logic. When the anomaly probability > 0.9 for three consecutive time steps, a data anomaly alarm is triggered.

[0112] 3. Use a random forest classifier for fault classification (such as internal short circuit or thermal runaway of the battery, etc.). The following further elaborates on the random forest classifier:

[0113] Feature engineering. Select the top 20 key features (such as voltage drop rate or temperature rise slope, etc.).

[0114] Model training. Hyperparameter tuning: Grid search to determine the optimal parameters (n_estimators = 200, max_depth = 12). Class weights. Set weights according to the frequency of occurrence of fault classes.

[0115] Decision explanation. Analyze the feature contribution degree through the values (Shapley Additive ex Planation, SHAP) used to explain the model prediction results to assist manual review.

[0116] 4. Generate maintenance suggestions and diagnostic reports in combination with the expert knowledge base.

[0117] Rule engine. For example, if the temperature > 60°C and the voltage fluctuation > 10%, the diagnostic result is thermal runaway.

[0118] Report generation. Generate a structured report through template filling technology. The structured report can be in the format of lightweight data exchange (JavaScript Object Notation, JSON). Generate natural language maintenance suggestions, such as "It is recommended to immediately stop charging and check the battery module connector", etc.

[0119] Based on the technical solution of the present invention, centralized acquisition of data, fault monitoring, fault classification, dynamic optimization of device operation strategies, and remote operation and maintenance of distributed energy storage devices can be achieved.

[0120] Embodiment 4

[0121] Figure 6 It is a schematic structural diagram of a monitoring device for an energy storage device provided in Embodiment 4 of the present invention. As Figure 6 shown, the device includes: a distributed device determination module 410, a device fault detection module 420, and a monitoring result determination module 430.

[0122] Among them, the distributed device determination module 410 is configured to determine a distributed device, where the distributed device includes a plurality of first energy storage devices; the device failure detection module 420 is configured to, for each of the first energy storage devices, determine real-time energy storage timing data of the first energy storage device, predict the input real-time energy storage timing data through a first prediction model to obtain predicted energy storage timing data, and determine a failure detection result of the first energy storage device according to the predicted energy storage timing data, where the first prediction model is obtained by training a neural network with a first training sample; the monitoring result determination module 430 is configured to determine a monitoring result of the distributed device according to a plurality of the failure detection results.

[0123] The technical solution of the embodiment of the present invention determines a distributed device, where the distributed device includes a plurality of first energy storage devices; for each of the first energy storage devices, determines real-time energy storage timing data of the first energy storage device, predicts the input real-time energy storage timing data through a first prediction model to obtain predicted energy storage timing data, and determines a failure detection result of the first energy storage device according to the predicted energy storage timing data, where the first prediction model is obtained by training a neural network with a first training sample; determines a monitoring result of the distributed device according to a plurality of the failure detection results. The present invention achieves the effect of performing high-accuracy centralized failure detection on multiple distributed energy storage devices, and realizes high-quality unified monitoring of multiple distributed energy storage devices by a cloud monitoring terminal.

[0124] Optionally, the device failure detection module 420 includes: an energy storage data prediction unit and a predicted timing determination unit;

[0125] Among them, the energy storage data prediction unit is configured to perform energy storage data prediction on the input real-time energy storage timing data through the first prediction model to obtain predicted energy storage data;

[0126] The predicted timing determination unit is configured to determine predicted energy storage timing data according to the predicted energy storage data and the real-time energy storage timing data.

[0127] Optionally, the predicted energy storage timing data includes the predicted energy storage data, real-time energy storage data, and reference energy storage sequence data;

[0128] The device failure detection module 420 includes: a failure detection unit, a detection result retrieval unit, and a detection result determination unit;

[0129] Among them, the failure detection unit is configured to perform anomaly detection on the real-time energy storage data to obtain a first detection result; and perform anomaly detection on the predicted energy storage data to obtain a second detection result;

[0130] The detection result retrieval unit is configured to determine a third detection result of reference energy storage data adjacent to the real-time energy storage data in time sequence in the reference energy storage sequence data;

[0131] The detection result determination unit is configured to determine a fault detection result of the first energy storage device according to the first detection result, the second detection result, and the third detection result.

[0132] Optionally, the fault detection result includes having a fault or not having a fault;

[0133] The monitoring device of the energy storage device further includes: a fault category identification module, configured to, after determining the fault detection result of the first energy storage device according to the predicted energy storage time sequence data, when the fault detection result is having a fault, determine energy storage abnormal data in the predicted energy storage time sequence data, and classify the input energy storage abnormal data through a first classification model to obtain a first fault category of the first energy storage device, where the first classification model is obtained by training a random forest model with a second training sample;

[0134] The monitoring result determination module 430 is specifically configured to: determine the monitoring result of the distributed device according to a plurality of the fault detection results and the first fault category.

[0135] Optionally, the fault category identification module includes: an abnormal data determination unit, configured to use the reference energy storage data, the real-time energy storage data, and the predicted energy storage data as energy storage abnormal data.

[0136] Optionally, the monitoring device of the energy storage device further includes: a maintenance suggestion determination module, configured to, after obtaining the first fault category of the first energy storage device, determine maintenance suggestion data corresponding to the first energy storage device with a fault through a pre-constructed category-maintenance mapping relationship library and the first fault category, where the category-maintenance mapping relationship library includes a variety of candidate fault categories, a variety of candidate maintenance suggestions, and a mapping relationship between the candidate fault categories and the candidate maintenance suggestions;

[0137] The monitoring result determination module 430 is specifically configured to: determine the monitoring result of the distributed device according to a plurality of the fault detection results, the first fault category, and the maintenance suggestion data.

[0138] Optionally, the device fault detection module 420 further includes: a real-time data acquisition unit and a reference sequence acquisition unit;

[0139] Among them, the real-time data acquisition unit is used to acquire the real-time energy storage data of the first energy storage device, where the energy storage data includes at least one of current, voltage, charging power, discharging power, temperature, state of charge, and health state;

[0140] The reference sequence acquisition unit is used to acquire the reference energy storage time series data corresponding to the first energy storage device, and determine the real-time energy storage time series data according to the reference energy storage time series data and the real-time energy storage data.

[0141] The monitoring device for the energy storage device provided by the embodiment of the present invention can execute the monitoring method for the energy storage device provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0142] Embodiment Five

[0143] Figure 7 The structural schematic diagram of the electronic device 10 that can be used to implement the embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0144] As Figure 7 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by at least one processor, and the processor 11 can execute various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0145] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0146] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the monitoring method of the energy storage device.

[0147] In some embodiments, the monitoring method of the energy storage device can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the monitoring method of the energy storage device described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the monitoring method of the energy storage device by any other suitable means (e.g., by means of firmware).

[0148] The various embodiments of the systems and technologies described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0149] A computer program for implementing the method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer programs are executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer programs can be executed entirely on the machine, partially on the machine, executed partially on the machine and partially on a remote machine as an independent software package, or executed entirely on a remote machine or server.

[0150] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0151] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0152] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected with each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0153] A computing system can include a client and a server. The client and the server are generally remote from each other and typically interact via a communication network. The client-server relationship is created by computer programs that run on respective computers and have a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0154] It should be understood that various forms of the processes shown above can be used, steps can be reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitations are imposed herein.

[0155] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A monitoring method for an energy storage device, characterized in that, Including: Determine distributed devices, where the distributed devices include multiple first energy storage devices; For each of the first energy storage devices, determine the real-time energy storage timing data of the first energy storage device, predict the input real-time energy storage timing data through a first prediction model to obtain predicted energy storage timing data, and determine the fault detection result of the first energy storage device according to the predicted energy storage timing data, where the first prediction model is obtained by training a neural network with a first training sample; Determine the monitoring result of the distributed device according to the multiple fault detection results.

2. The method according to claim 1, wherein The predicting the input real-time energy storage timing data through the first prediction model to obtain predicted energy storage timing data includes: Predict the energy storage data of the input real-time energy storage timing data through the first prediction model to obtain predicted energy storage data; Determine the predicted energy storage timing data according to the predicted energy storage data and the real-time energy storage timing data.

3. The method according to claim 2, wherein The predicted energy storage timing data includes the predicted energy storage data, the real-time energy storage data, and the reference energy storage sequence data; the determining the fault detection result of the first energy storage device according to the predicted energy storage timing data includes: Perform anomaly detection on the real-time energy storage data to obtain a first detection result; and perform anomaly detection on the predicted energy storage data to obtain a second detection result; Determine a third detection result of the reference energy storage data adjacent to the real-time energy storage data in time sequence in the reference energy storage sequence data; Determine the fault detection result of the first energy storage device according to the first detection result, the second detection result, and the third detection result.

4. The method according to claim 1, characterized in that, The fault detection result includes having a fault or not having a fault; After determining the fault detection result of the first energy storage device according to the predicted energy storage timing data, it further includes: In the case where the fault detection result is having a fault, determine the energy storage anomaly data in the predicted energy storage timing data, and classify the input energy storage anomaly data through a first classification model to obtain the first fault category of the first energy storage device, where the first classification model is obtained by training a random forest model with a second training sample; The determining the monitoring result of the distributed device according to the multiple fault detection results includes: Determine the monitoring result of the distributed device according to the multiple fault detection results and the first fault category.

5. The method according to claim 3 or 4, characterized in that, The determining the energy storage anomaly data in the predicted energy storage timing data includes: Use the reference energy storage data, the real-time energy storage data, and the predicted energy storage data as the energy storage anomaly data.

6. The method according to claim 4, wherein After obtaining the first fault category of the first energy storage device, it further includes: Determine the maintenance advice data corresponding to the first energy storage device with a fault through a pre-constructed category-maintenance mapping relationship library and the first fault category, where the category-maintenance mapping relationship library includes multiple candidate fault categories, multiple candidate maintenance advices, and the mapping relationship between the candidate fault categories and the candidate maintenance advices; Determining the monitoring result of the distributed device according to the multiple fault detection results includes: Determining the monitoring result of the distributed device according to the multiple fault detection results, the first fault category, and the maintenance suggestion data.

7. The method according to claim 1, characterized in that, Determining the real-time energy storage timing data of the first energy storage device includes: Obtaining the real-time energy storage data of the first energy storage device, where the energy storage data includes at least one of current, voltage, charging power, discharging power, temperature, state of charge, and health state; Obtaining the reference energy storage timing data corresponding to the first energy storage device, and determining the real-time energy storage timing data according to the reference energy storage timing data and the real-time energy storage data.

8. A monitoring device for an energy storage device, characterized in that, Including: A distributed device determination module for determining a distributed device, where the distributed device includes multiple first energy storage devices; A device fault detection module for, for each of the first energy storage devices, determining the real-time energy storage timing data of the first energy storage device, predicting the input real-time energy storage timing data through a first prediction model to obtain predicted energy storage timing data, and determining the fault detection result of the first energy storage device according to the predicted energy storage timing data, where the first prediction model is obtained by training a neural network with a first training sample; A monitoring result determination module for determining the monitoring result of the distributed device according to the multiple fault detection results.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; where The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the monitoring method of the energy storage device according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a processor to execute the monitoring method of the energy storage device according to any one of claims 1-7 when executed.

Images