A method, system, device and storage medium for detecting an energy storage power station
By employing non-intrusive data acquisition, universal message parsing, and GPU-optimized computation, the problems of inconsistent data formats and low computational efficiency in energy storage power station testing have been solved, enabling the efficient and accurate operation of the portable testing system.
Patent Information
- Application Number
- CN202610675275.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-05-15
- Publication Date
- 2026-08-25
AI Technical Summary
The testing of energy storage power stations suffers from problems such as intrusive data acquisition, inconsistent equipment data formats, low computational efficiency, limited testing system functionality, and poor compatibility, making it difficult to meet the needs for rapid and accurate on-site testing.
Employing a non-intrusive data acquisition method, a general-purpose message parsing engine, and massive data acceleration computing technology, combined with GPU-optimized model compilation, it achieves standardized data processing and efficient computing, integrating a portable live-line detection system.
It enables plug-and-play and rapid deployment, improves data integration efficiency and computing speed, reduces system costs, and enhances the convenience and accuracy of detection.
Smart Images
Figure CN122634036A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of energy storage power station testing technology, and particularly relates to an energy storage power station testing method, system, equipment and storage medium. Background Technology
[0002] In the operation and maintenance of energy storage power stations, testing is a crucial step in ensuring the safe and stable operation of the station. However, there are currently many technical challenges in the field of energy storage power station testing: In terms of data acquisition, there is a lack of universal, non-intrusive, rapid data acquisition technology; existing acquisition methods often require modification or interruption of energy storage equipment, affecting normal equipment operation, and the data formats and protocols of different manufacturers and models of energy storage equipment are not uniform, making data standardization processing difficult and hindering efficient data integration and analysis.
[0003] In terms of data computation, in portable energy storage testing scenarios, it is difficult to balance computational efficiency and accuracy for massive amounts of testing data. Traditional computing solutions cannot adapt to the characteristics of energy storage testing algorithm models. Inappropriate operator scheduling and lack of hardware optimization in model compilation result in slow computation speed, failing to meet the needs of rapid on-site testing.
[0004] In terms of testing systems, there is a lack of integrated portable live-line testing systems. Existing testing equipment is mostly single-function equipment with poor software and hardware compatibility, high cost, and insufficient scalability and maintainability, making it difficult to adapt to complex on-site testing environments. Summary of the Invention
[0005] This invention provides a method, system, device, and storage medium for testing energy storage power stations, aiming to solve the technical problems existing in the prior art, such as the intrusive nature of data acquisition for energy storage power stations, which affects the normal operation of equipment, and the inconsistency of data formats and protocols among different devices, making standardization difficult; low computational efficiency for massive amounts of data in portable energy storage testing scenarios, poor operator-hardware compatibility, unoptimized model compilation, and inability to balance efficiency and accuracy; and the lack of integrated portable live-line testing systems, with single device functions and poor compatibility, scalability, and maintainability.
[0006] The technical solution of this invention to solve the above-mentioned technical problems is as follows: A detection method for an energy storage power station, comprising: The energy storage power station equipment is used to collect data in real time using a non-intrusive data acquisition method to obtain the collected data; The collected data is parsed using a general-purpose energy storage device message parsing engine to obtain parsed data; The parsed data is standardized to obtain standardized data, and the standardized data is stored in a relational database. We apply massive data acceleration computing methods to perform in-depth analysis on standardized data in relational databases and obtain in-depth analysis results. A detection report is generated based on the in-depth analysis results, and the detection report is then visualized.
[0007] Furthermore, the aforementioned non-intrusive data acquisition method includes: capturing communication messages of energy storage power station equipment through a port monitoring module in a bypass listening manner; wherein the acquisition frequency is adjustable during the data acquisition process.
[0008] Furthermore, the aforementioned general-purpose energy storage device message parsing engine has a built-in protocol feature library and parsing rule library; the collected data is parsed using the general-purpose energy storage device message parsing engine to obtain the parsed data as follows: The protocol identifier and data frame structure of the collected data are obtained, and the protocol type of the collected data is obtained by matching based on the protocol feature library. Based on the aforementioned parsing rule base, the collected data is parsed according to the parsing rules corresponding to the protocol type of the collected data to obtain parsed data.
[0009] Furthermore, the above-mentioned standardization processing of the parsed data includes: converting the data structure and original measurement point codes of energy storage power station equipment of different specifications according to a preset unified data standard.
[0010] Furthermore, the aforementioned methods for accelerating computation on massive amounts of data include: Analyze the computational requirements of each energy storage detection model; Based on the aforementioned computational requirements, dedicated operators are deployed on the GPU, including voltage analysis operators and temperature field calculation operators; Based on the computational complexity of the dedicated operator and the dependency relationship of the standardized data, a dynamic scheduling method is used to allocate GPU resources for each parallel-running energy storage detection model.
[0011] Furthermore, prior to deploying dedicated operators on GPUs, the aforementioned method for accelerating massive data computation also includes: The GPU resource utilization rate during the operation of each energy storage detection model and the interaction relationship between the models are analyzed to obtain the model analysis results. Based on the model analysis results, a model priority evaluation mechanism is established to classify the multiple energy storage detection models into a first type and a second type. The first type of energy storage detection model is compiled and optimized using an instruction-level optimization method, while the second type of energy storage detection model is compiled and optimized using a lightweight compilation method. Based on the GPU's hardware performance parameters, optimize memory access patterns and thread allocation strategies during compilation; The compiled and optimized energy storage detection models are deployed collaboratively in the same GPU execution environment, and multiple models can run in parallel through unified scheduling.
[0012] Furthermore, the above analysis of the GPU resource occupancy rate during the operation of each energy storage detection model, and the interaction relationship between each energy storage detection model, also includes: training the energy storage detection model to obtain a trained energy storage detection model, specifically: Acquire historical running data, preprocess the historical running data, and obtain a training dataset; The training dataset is input into the energy storage detection model, and the energy storage detection model is iteratively trained using the mini-batch gradient descent method until the model converges or reaches the preset number of training rounds, thus obtaining the trained energy storage detection model.
[0013] Secondly, to solve the above-mentioned technical problems, the present invention also provides an energy storage power station detection system, comprising: The data acquisition module is used to acquire real-time data from energy storage power station equipment using non-intrusive data acquisition methods to obtain the acquired data. The data parsing module is used to parse the collected data using a general energy storage device message parsing engine to obtain parsed data; The standardization module is used to standardize the parsed data to obtain standardized data, and store the standardized data in a relational database. The deep analysis module is used to perform deep analysis on standardized data in relational databases using massive data acceleration computing methods to obtain deep analysis results; The visualization module is used to generate a test report based on the in-depth analysis results and to visualize the test report.
[0014] Thirdly, in order to solve the above-mentioned technical problems, the present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the energy storage power station detection method of the present application.
[0015] Fourthly, in order to solve the above-mentioned technical problems, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the energy storage power station detection method of the present application.
[0016] Compared with the prior art, the present invention has the following advantages: 1. This invention employs a bypass monitoring method with a universal non-intrusive data rapid acquisition technology, avoiding interference with the normal operation of energy storage devices and enabling plug-and-play rapid deployment; through universal message parsing and data reconstruction technology, it solves the data standardization problem of devices from different manufacturers and models, and improves data integration efficiency.
[0017] 2. The massive data acceleration computing technology of this invention fully leverages the computing performance of GPUs through the development of dedicated operators and optimization of model compilation, significantly improving the computing speed of massive energy storage data; combined with model training and optimization based on actual operating data, it ensures computing accuracy and meets the efficiency and accuracy requirements of portable detection scenarios.
[0018] 3. The portable live-line detection system of the present invention integrates data acquisition, calculation and analysis and display functions. It is small in size and has a long battery life, making it convenient for on-site live-line detection. Its modular architecture design gives it good scalability and maintainability, reduces system costs and improves the convenience and efficiency of detection work.
[0019] Other features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description
[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0021] Figure 1 A flowchart illustrating a detection method for an energy storage power station according to an embodiment of the present invention is shown; Figure 2 A schematic diagram of the structure of an energy storage power station detection system according to an embodiment of the present invention is shown; Figure 3 A schematic diagram of an electronic device structure according to an embodiment of the present invention is shown. Detailed Implementation
[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0023] Figure 1 A flowchart illustrating a testing method for an energy storage power station according to an embodiment of the present invention is shown, as follows: Figure 1 As shown, an embodiment of the present invention provides a method for testing an energy storage power station, comprising: The energy storage power station equipment is used to collect data in real time using a non-intrusive data acquisition method to obtain the collected data; The collected data is parsed using a general-purpose energy storage device message parsing engine to obtain parsed data; The parsed data is standardized to obtain standardized data, and the standardized data is stored in a relational database. We apply massive data acceleration computing methods to perform in-depth analysis on standardized data in relational databases and obtain in-depth analysis results. A detection report is generated based on the in-depth analysis results, and the detection report is then visualized.
[0024] In this embodiment, the collected data includes: voltage data, current data, temperature data, and operating status data; wherein, the operating status data includes: SOC, charge / discharge status data, and alarm data.
[0025] Optionally, the non-intrusive data acquisition method includes: capturing communication messages of the energy storage power station equipment through a port monitoring module in a bypass listening mode; wherein the acquisition frequency is adjustable during the data acquisition process.
[0026] In this embodiment, the port monitoring module is plug-and-play and can be quickly configured through a preset configuration template. The module has a built-in packet capture unit, which can achieve non-intrusive high-frequency packet acquisition by using a bypass listening method without interfering with the normal packet processing flow of the device. The acquisition frequency supports up to once every 1 millisecond, and does not interfere with the original packet processing flow of the device.
[0027] Optionally, the general-purpose energy storage device message parsing engine has a built-in protocol feature library and parsing rule library; the collected data is parsed using the general-purpose energy storage device message parsing engine to obtain the parsed data as follows: The protocol identifier and data frame structure of the collected data are obtained, and the protocol type of the collected data is obtained by matching based on the protocol feature library. Based on the aforementioned parsing rule base, the collected data is parsed according to the parsing rules corresponding to the protocol type of the collected data to obtain parsed data.
[0028] In this embodiment, the protocol type includes Modbus / TCP, IEC104 or IEC61850 protocol. After parsing the collected data into structured data, it is mapped to the measurement point fields corresponding to the entire database.
[0029] Optionally, standardizing the parsed data includes converting the data structure and original measurement point codes of energy storage power station equipment of different specifications according to a preset unified data standard.
[0030] In this embodiment, by standardizing energy storage power station equipment from different manufacturers, data types are unified and coding rules are adapted. A mapping relationship is established between database measurement point fields and front-end visualization variables. After the standardized data is calculated by the built-in data processing module, the calculation results are quickly displayed through the front-end visualization component.
[0031] Optionally, the method for accelerating computation on massive amounts of data includes: Analyze the computational requirements of each energy storage detection model; Based on the aforementioned computational requirements, dedicated operators are deployed on the GPU, including voltage analysis operators and temperature field calculation operators; Based on the computational complexity of the dedicated operator and the dependency relationship of the standardized data, a dynamic scheduling method is used to allocate GPU resources for each parallel-running energy storage detection model.
[0032] In this embodiment, the energy storage detection model includes a battery state assessment model and a fault diagnosis model; the parallel computing efficiency is improved by efficiently allocating GPU resources through a dynamic scheduling method.
[0033] In this embodiment, an NVIDIA Jetson AGX Orin GPU is selected as the computing core. The voltage analysis operator is used to perform consistency assessment of the battery voltage sequence, anomaly detection, and health status feature extraction. The temperature field calculation operator is used to calculate the spatial gradient of multi-point temperatures, temperature rise rate, or identification of thermal runaway risk areas. The voltage analysis and temperature field calculation operators are high-performance computing units specifically designed for energy storage detection tasks. The development process of these computing units includes: analyzing the computational requirements of the energy storage detection model. Subsequently, these specific computational requirements are transformed into structured computational tasks and further abstracted into reusable and parameterizable computational patterns. For example, "sliding window range / standard deviation calculation of multi-channel time-series signals" is refined into the basic paradigm of the voltage analysis operator, while "estimation of local temperature gradient and temperature rise rate on a three-dimensional topological mesh" constitutes the core logic of the temperature field calculation operator. Based on this, and combined with the hardware architecture characteristics of the target GPU, the operators are implemented at the underlying level using CUDA or HIP programming models. Finally, the optimized GPU kernel is encapsulated into a callable operator interface.
[0034] Optionally, before deploying dedicated operators on the GPU, the method for accelerating computation on massive amounts of data also includes: The GPU resource utilization rate during the operation of each energy storage detection model and the interaction relationship between the models are analyzed to obtain the model analysis results. Based on the model analysis results, a model priority evaluation mechanism is established to classify the multiple energy storage detection models into a first type and a second type. The first type of energy storage detection model is compiled and optimized using an instruction-level optimization method, while the second type of energy storage detection model is compiled and optimized using a lightweight compilation method. Based on the GPU's hardware performance parameters, optimize memory access patterns and thread allocation strategies during compilation; The compiled and optimized energy storage detection models are deployed collaboratively in the same GPU execution environment, and multiple models can run in parallel through unified scheduling.
[0035] In this embodiment, the instruction-level optimization includes at least one of loop unrolling, memory access rearrangement, or register allocation optimization to improve its execution efficiency on the GPU; the lightweight compilation includes at least one of model pruning, parameter quantization, or operator fusion to reduce its computational overhead while ensuring overall detection accuracy; the hardware performance parameters include at least one of the number of stream processors, memory bandwidth, or cache structure.
[0036] Optionally, analyzing the GPU resource occupancy rate of each energy storage detection model during operation, and the interaction relationships between each energy storage detection model, further includes: training the energy storage detection model to obtain a trained energy storage detection model, specifically: Acquire historical running data, preprocess the historical running data, and obtain a training dataset; The training dataset is input into the energy storage detection model, and the energy storage detection model is iteratively trained using the mini-batch gradient descent method until the model converges or reaches the preset number of training rounds, thus obtaining the trained energy storage detection model.
[0037] In this embodiment, historical data includes voltage data, current data, temperature data, and operating status data; among which, operating status data includes SOC, charge / discharge status data, and alarm data. Specifically, it includes historical operating data of operational energy storage power stations (including battery voltage, current, temperature, etc. under normal and fault conditions). The energy storage detection model is not a single model, but a set of algorithm models for different detection targets, including: battery status assessment model, fault diagnosis model, and anomaly detection model.
[0038] In this embodiment, the energy storage detection algorithm model is trained using the mini-batch gradient descent method. The test dataset (containing samples with known detection results) is input into the trained model, and the model's performance is evaluated using metrics such as accuracy and recall. The data samples to be detected are then input into the model for inference, and the model parameters are adjusted based on the deviation between the inference results and the actual detection results.
[0039] Based on and Figure 1 Using the same principle as the method shown, this embodiment of the invention also provides an energy storage power station detection system, such as... Figure 2 As shown, it includes: The data acquisition module is used to acquire real-time data from energy storage power station equipment using non-intrusive data acquisition methods to obtain the acquired data. The data parsing module is used to parse the collected data using a general energy storage device message parsing engine to obtain parsed data; The standardization module is used to standardize the parsed data to obtain standardized data, and store the standardized data in a relational database. The deep analysis module is used to perform deep analysis on standardized data in relational databases using massive data acceleration computing methods to obtain deep analysis results; The visualization module is used to generate a test report based on the in-depth analysis results and to visualize the test report.
[0040] In this embodiment, the data acquisition module adopts a USB interface design with a built-in automatic identification chip. After being connected to the communication port of the energy storage device, it can complete the configuration within 30 seconds. The default acquisition frequency is set to 50ms / time, which can be adjusted to the range of 10-1000ms through the software interface. The message parsing engine has 100+ built-in protocol features, and the parsing time for Modbus / TCP protocol messages is low (≤10ms) with high parsing accuracy.
[0041] The standardization module processes a single piece of device data in ≤5ms, and the converted standardized data is stored in a MySQL database. After the mapping relationship of the automatic mapping engine is configured, the latency of updating the database data to the front-end display is very low (≤200ms), and the front-end uses WebGL technology to realize the real-time rendering of data curves.
[0042] In the deep analysis module, the NVIDIA Jetson AGX Orin GPU was selected as the computing core. The developed voltage analysis operator achieved improved computational efficiency compared to general-purpose operators, and the parallelism of the temperature field calculation operator was also improved. The operator scheduling algorithm can dynamically adjust the allocation strategy according to the real-time computing load, thereby improving the utilization of GPU resources.
[0043] After model compilation optimization, the efficiency of multi-model collaborative operation was improved. Training the model with 100,000 historical data points improved the model's accuracy on the test set and reduced the time required for inferring a single data point.
[0044] The energy storage power station testing system of this invention can execute the energy storage power station testing method provided in this invention. The implementation principle is similar. The actions performed by each module and unit in the energy storage power station testing system of each embodiment of this invention correspond to the steps in the energy storage power station testing method of each embodiment of this invention. For detailed functional descriptions of each module of the energy storage power station testing system, please refer to the descriptions in the corresponding energy storage power station testing methods shown above, which will not be repeated here.
[0045] The aforementioned energy storage power station testing system can be a computer program (including program code) running on a computer device, such as an application software; the application software can be used to execute the corresponding steps in the method provided in the embodiments of the present invention.
[0046] The modules described in the embodiments of the present invention can be implemented in software or hardware. The names of the modules are not, in some cases, limiting the scope of the module itself.
[0047] Based on the same principles as the methods shown in the embodiments of the present invention, the embodiments of the present invention also provide an electronic device, which may include, but is not limited to: a processor and a memory; the memory for storing computer programs; and the processor for executing the methods shown in any embodiment of the present invention by invoking the computer programs.
[0048] In one alternative embodiment, an electronic device is provided, such as Figure 3 As shown, Figure 3 The illustrated electronic device includes a processor and a memory. The processor and memory are connected, for example, via a bus. Optionally, the electronic device may also include a transceiver, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver is not limited to one unit, and the structure of this electronic device does not constitute a limitation on the embodiments of the present invention.
[0049] The memory stores application code (computer program) that executes the present invention, and its execution is controlled by a processor. The processor executes the application code stored in the memory to implement the content shown in the foregoing method embodiments.
[0050] Among these, electronic devices can also be terminal devices. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention.
[0051] This invention provides a computer-readable storage medium storing a computer program that, when run on a computer, enables the computer to execute the corresponding content in the aforementioned method embodiments.
[0052] According to another aspect of the present invention, a computer program product or computer program is also provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various embodiments described above.
[0053] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.
[0054] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.
Claims
1. A testing method for an energy storage power station, characterized in that, include: The energy storage power station equipment is used to collect data in real time using a non-intrusive data acquisition method to obtain the collected data; The collected data is parsed using a general-purpose energy storage device message parsing engine to obtain parsed data; The parsed data is standardized to obtain standardized data, and the standardized data is stored in a relational database. We apply massive data acceleration computing methods to perform in-depth analysis on standardized data in relational databases and obtain in-depth analysis results. A detection report is generated based on the in-depth analysis results, and the detection report is then visualized.
2. The energy storage power station testing method according to claim 1, characterized in that, The non-intrusive data acquisition method includes: capturing communication messages of energy storage power station equipment through a port monitoring module in a bypass listening mode; wherein the acquisition frequency is adjustable during the data acquisition process.
3. The energy storage power station testing method according to claim 1, characterized in that, The general-purpose energy storage device message parsing engine has a built-in protocol feature library and parsing rule library; the collected data is parsed using the general-purpose energy storage device message parsing engine to obtain the parsed data as follows: The protocol identifier and data frame structure of the collected data are obtained, and the protocol type of the collected data is obtained by matching based on the protocol feature library. Based on the aforementioned parsing rule base, the collected data is parsed according to the parsing rules corresponding to the protocol type of the collected data to obtain parsed data.
4. The energy storage power station testing method according to claim 1, characterized in that, Standardizing the parsed data includes converting the data structure and original measurement point codes of energy storage power station equipment of different specifications according to a preset unified data standard.
5. The energy storage power station testing method according to claim 1, characterized in that, The method for accelerating computation of massive amounts of data includes: Analyze the computational requirements of each energy storage detection model; Based on the aforementioned computational requirements, dedicated operators are deployed on the GPU, including voltage analysis operators and temperature field calculation operators; Based on the computational complexity of the dedicated operator and the dependency relationship of the standardized data, a dynamic scheduling method is used to allocate GPU resources for each parallel-running energy storage detection model.
6. The energy storage power station testing method according to claim 5, characterized in that, Prior to deploying dedicated operators on GPUs, the method for accelerating computation on massive amounts of data also includes: The GPU resource utilization rate during the operation of each energy storage detection model and the interaction relationship between the models are analyzed to obtain the model analysis results. Based on the model analysis results, a model priority evaluation mechanism is established to classify the multiple energy storage detection models into a first type and a second type. The first type of energy storage detection model is compiled and optimized using an instruction-level optimization method, while the second type of energy storage detection model is compiled and optimized using a lightweight compilation method. Based on the GPU's hardware performance parameters, optimize memory access patterns and thread allocation strategies during compilation; The compiled and optimized energy storage detection models are deployed collaboratively in the same GPU execution environment, and multiple models can run in parallel through unified scheduling.
7. The energy storage power station testing method according to claim 6, characterized in that, The analysis of GPU resource occupancy during the operation of each energy storage detection model, and the interaction relationships between them, also includes: training the energy storage detection models to obtain trained energy storage detection models, specifically: Acquire historical running data, preprocess the historical running data, and obtain a training dataset; The training dataset is input into the energy storage detection model, and the energy storage detection model is iteratively trained using the mini-batch gradient descent method until the model converges or reaches the preset number of training rounds, thus obtaining the trained energy storage detection model.
8. A detection system for an energy storage power station, characterized in that, include: The data acquisition module is used to acquire real-time data from energy storage power station equipment using non-intrusive data acquisition methods to obtain the acquired data. The data parsing module is used to parse the collected data using a general energy storage device message parsing engine to obtain parsed data; The standardization module is used to standardize the parsed data to obtain standardized data, and store the standardized data in a relational database. The deep analysis module is used to perform deep analysis on standardized data in relational databases using massive data acceleration computing methods to obtain deep analysis results; The visualization module is used to generate a test report based on the in-depth analysis results and to visualize the test report.
9. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method of any one of claims 1-7.
10. A computer storage medium, characterized in that, The computer storage medium stores a computer program, which, when executed by a processor, implements the method of any one of claims 1-7.