CN-122015959-A - Electrochemical energy storage system safety assessment method integrating multi-sensor data

CN122015959ACN 122015959 ACN122015959 ACN 122015959ACN-122015959-A

Abstract

The invention discloses a safety assessment method of an electrochemical energy storage system for fusing multi-sensor data, which relates to the technical field of electrochemical energy storage safety monitoring and comprises the following steps of S1, multi-mode data acquisition, S2, data preprocessing, S3, multi-sensor data fusion, weighting feature fusion of standardized data obtained in the step S2 through a multi-sensor fusion algorithm built in a microcontroller, outputting fusion feature values, S4, safety state assessment, and S5, early warning and alarm and communication. According to the invention, through deployment of various sensing devices such as a gas sensor array, a high-precision temperature sensor group, a current-voltage Hall sensor and the like, multi-dimensional data acquisition of the running state of the electrochemical energy storage system is realized, and a complete energy storage system safety monitoring system is constructed by matching with synchronous acquisition and fusion analysis of multi-source data, so that monitoring blind areas existing in a single sensor are effectively overcome, and early warning and accurate identification of potential risks such as thermal runaway, overcharge and overdischarge are realized.

Inventors

HE HAO
HE YUTAO
HUANG DONGMEI
LI CONGCONG
WEN ZHENYU
Nian Mingming

Assignees

中国计量大学

Dates

Publication Date: 20260512
Application Date: 20260130

Claims (10)

1. The electrochemical energy storage system safety assessment method integrating the multi-sensor data is characterized by comprising the following steps of: S1, multi-mode data acquisition, namely using an STM32F103VET6 microcontroller as a control core to drive a CO sensor, a VOC gas sensor, The gas sensor, the temperature sensor and the voltage and current sensor synchronously acquire gas concentration data, temperature data and charge and discharge loop voltage and current data of the electrochemical energy storage system according to a preset sampling frequency; S2, data preprocessing, namely performing filtering denoising, outlier rejection and standardization processing on the acquired multi-mode original data by an STM32F103VET6 microcontroller to obtain standardized data; S3, multi-sensor data fusion, namely calculating the variance contribution rate of each sensor data in real time by using a multi-sensor fusion algorithm built in a microcontroller, and solving the fusion distortion problem of the traditional fixed weight when the SOC is more than or equal to 80% or less than or equal to 20%: Calculating the measurement error variance sigma i2 of each sensor based on the deviation between the real-time acquisition data of the sensor and a historical reference database; step two, dynamically distributing weight coefficients, namely calculating the weight coefficients of the sensors according to the error variance, and following the principle that the smaller the error is, the larger the weight is; step three, weighting summation preliminary fusion, namely weighting summation operation is carried out on the standardized data obtained in the step S2 by utilizing the dynamic weight coefficient to obtain a preliminary fusion result And step four, main component extraction optimization, namely performing feature extraction on the primary fusion result by adopting a main component analysis (PCA) technology, removing redundant information, retaining key features and finally obtaining a fusion feature value. S4, safety state assessment, namely presetting an electrochemical energy storage system safety assessment index system comprising a gas concentration threshold value interval, a temperature threshold value interval and a voltage current rated range, comparing the fusion characteristic value in the step S3 with the assessment index system, outputting a safety state grade, and based on the fusion characteristic value, adopting an energy storage system exclusive risk matrix and referencing a nuclear power grade RMAP standard to divide the safety state into a unacceptable characteristic value A, a characteristic value more than or equal to 80, a characteristic value less than or equal to 40 to be relieved B <80, and a characteristic value acceptable C <40, wherein the A grade corresponds to a precursor of thermal runaway of the battery cell, and the B grade corresponds to leakage/local overheating of electrolyte; and S5, early warning and communication, namely triggering corresponding acousto-optic early warning/alarm signals according to the safety state level of the step S4, uploading real-time state data, fusion characteristic values and safety level information to a monitoring platform through a communication module, enabling the early warning signals to be communicated with an energy storage BMS in real time through a CAN bus, enabling a class A risk to trigger an acousto-optic warning and an emergency stop instruction, and enabling a class B risk to trigger short message pushing and operation and maintenance work order generation.
2. The method for evaluating the safety of the electrochemical energy storage system fused with the multi-sensor data according to claim 1, wherein in the step S1, the preset sampling frequency is 10-100 Hz, and the STM32F103VET6 microcontroller establishes communication with each sensor through an I2C, SPI or ADC interface to realize synchronous data acquisition.
3. The method for evaluating the safety of the electrochemical energy storage system fusing the multi-sensor data, as set forth in claim 1, is characterized in that in the step S1, for the electrochemical energy storage system in a high-temperature and high-humidity environment, the sensor is selected from the group consisting of a CO sensor which is an MQ-7H type high-temperature resistant electrochemical sensor, a VOC sensor which is a TGS2600-HT type high-humidity adaptive semiconductor sensor, and a temperature sensor which is a DS18B20-PRO type industrial grade digital sensor.
4. The method for evaluating the safety of the electrochemical energy storage system fused with the multi-sensor data according to claim 1, wherein in the step S2, the filtering denoising adopts a moving average filtering algorithm, the outlier rejection adopts a 3 sigma criterion, and the normalization processing adopts a Z-score normalization formula: wherein As the raw data is to be processed, As a mean value of the data, Data standard deviation.
5. The method for evaluating the safety of an electrochemical energy storage system fused with multi-sensor data according to claim 1, wherein in the step S3, the weight coefficient of the improved weighted least squares fusion algorithm is calculated by: wherein For the weight coefficient of the i-th sensor, The measurement error variance of the ith sensor is given, and n is the number of sensors.
6. The method for evaluating the safety of the electrochemical energy storage system fusing the multi-sensor data according to claim 1, wherein in the step S3, the multi-sensor fusion algorithm further comprises a sensor real-time reliability calibration step, wherein each interval is 5-15 min, the STM32F103VET6 microcontroller calls a built-in historical reference database, and the deviation calculation is carried out on the current measured value of each sensor and the corresponding reference value.
7. The method for evaluating the safety of the electrochemical energy storage system fusing the multi-sensor data according to claim 1, wherein in the step S4, the safety evaluation index system is determined by combining rated operation parameters, fault evolution rules and industry safety standards of the electrochemical energy storage system, presetting a safety evaluation index system based on a analytic hierarchy process, and comparing fusion characteristic values with preset threshold intervals: If the fusion characteristic value falls in the normal level interval, judging that the system operates normally; if the potential safety risk falls in the early warning level interval, judging that the potential safety risk exists in the system; If the system falls in the alarm level interval, judging that the system has faults. The evaluation period is less than or equal to 100ms, and real-time dynamic evaluation is realized.
8. The method for evaluating the safety of the electrochemical energy storage system fusing the multi-sensor data according to claim 1, wherein in the step S4, a real-time dynamic update mechanism is adopted for evaluating the safety state, 1 fusing and evaluating are performed once 1 group of data is acquired, and the evaluating period is less than or equal to 100ms.
9. The method for evaluating the safety of the electrochemical energy storage system fused with the multi-sensor data according to claim 1, wherein in the step S5, the communication module is an RS485, loRa or Ethernet module, and a communication protocol adopts Modbus-RTU or MQTT to realize wired/wireless remote transmission of the data.
10. The method for evaluating the safety of the electrochemical energy storage system by fusing the multi-sensor data according to claim 1, wherein in the step S5, an early warning and alarming strategy is linked with an operation mode of the energy storage system: (1) In the charging mode, if the early warning is triggered, the charging current is preferentially reduced to 50% of the rated value and kept, if the warning is triggered, the charging loop is immediately cut off, and meanwhile, the module cooling fan is started; (2) In the discharging mode, if the early warning is triggered, the current discharging current is maintained but the duration of continuous discharging is limited to be less than or equal to 30min; (3) In the standby mode, if the early warning/alarming is triggered, except for the acousto-optic prompt, a 'device wake-up request' is sent to the monitoring platform through the communication module, and the protection action is executed after the platform confirms.

Description

Electrochemical energy storage system safety assessment method integrating multi-sensor data Technical Field The invention relates to the technical field of electrochemical energy storage safety monitoring, in particular to a safety assessment method of an electrochemical energy storage system fusing multi-sensor data. Background The electrochemical energy storage system is widely applied to the scenes of new energy power generation, power grid peak shaving and the like due to the advantages of high energy density, high response speed and the like. However, the electrochemical energy storage system is easy to cause safety accidents due to the problems of over-charge and over-discharge, thermal runaway, battery aging and the like in the charge and discharge process, and the characteristics of gas leakage, temperature rise, abnormal voltage and current and the like can be accompanied in the fault evolution process. The single sensor is easily affected by factors such as electromagnetic interference, temperature and humidity fluctuation and the like in a complex environment, so that larger deviation exists in measured data, and false alarm occurrence frequency in the system safety evaluation process are further caused. Disclosure of Invention The invention aims to provide a safety evaluation method of an electrochemical energy storage system fusing multi-sensor data, which aims to solve the problems in the background technology. In order to solve the technical problems, the invention adopts the following technical scheme: a safety assessment method of an electrochemical energy storage system integrating multi-sensor data comprises the following steps: S1, multi-mode data acquisition, namely using an STM32F103VET6 microcontroller as a control core to drive a CO sensor, a VOC gas sensor, The gas sensor, the temperature sensor and the voltage and current sensor synchronously acquire gas concentration data, temperature data and charge and discharge loop voltage and current data of the electrochemical energy storage system according to a preset sampling frequency; S2, data preprocessing, namely performing filtering denoising, outlier rejection and standardization processing on the acquired multi-mode original data by an STM32F103VET6 microcontroller to obtain standardized data; S3, multi-sensor data fusion, namely calculating the variance contribution rate of each sensor data in real time by using a multi-sensor fusion algorithm built in a microcontroller, and solving the fusion distortion problem of the traditional fixed weight when the SOC is more than or equal to 80% or less than or equal to 20%: Calculating the measurement error variance sigma i2 of each sensor based on the deviation between the real-time acquisition data of the sensor and a historical reference database; step two, dynamically distributing weight coefficients, namely calculating the weight coefficients of the sensors according to the error variance, and following the principle that the smaller the error is, the larger the weight is; step three, weighting summation preliminary fusion, namely weighting summation operation is carried out on the standardized data obtained in the step S2 by utilizing the dynamic weight coefficient to obtain a preliminary fusion result And step four, main component extraction optimization, namely performing feature extraction on the primary fusion result by adopting a main component analysis (PCA) technology, removing redundant information, retaining key features and finally obtaining a fusion feature value. S4, safety state assessment, namely presetting an electrochemical energy storage system safety assessment index system comprising a gas concentration threshold value interval, a temperature threshold value interval and a voltage current rated range, comparing the fusion characteristic value in the step S3 with the assessment index system, outputting a safety state grade, and based on the fusion characteristic value, adopting an energy storage system exclusive risk matrix and referencing a nuclear power grade RMAP standard to divide the safety state into a unacceptable characteristic value A, a characteristic value more than or equal to 80, a characteristic value less than or equal to 40 to be relieved B <80, and a characteristic value acceptable C <40, wherein the A grade corresponds to a precursor of thermal runaway of the battery cell, and the B grade corresponds to leakage/local overheating of electrolyte; and S5, early warning and communication, namely triggering corresponding acousto-optic early warning/alarm signals according to the safety state level of the step S4, uploading real-time state data, fusion characteristic values and safety level information to a monitoring platform through a communication module, enabling the early warning signals to be communicated with an energy storage BMS in real time through a CAN bus, enabling a class A risk to trigger an acousto-optic warning and an emergency stop inst