KR20220006799A - Smart battery management system for fire prevention - Google Patents

Abstract

Provided is a smart battery management system (S-BMS). According to an embodiment of the present invention, the S-BMS for integrally managing one or more batteries comprises: a battery state estimating module for estimating a state-of-charge (SOC) of a battery using a hybrid Kalman filter (HKF), and predicting R and C values of the battery which are state variables input to the HKF using a deep learning technique; a fire prevention information collecting module for collecting the temperature of a cell of each battery, a fire safety check item signal, and a relay contact signal; and a power control module for controlling charge/discharge of the battery by synthesizing signals from the battery state estimating module and the fire prevention information collecting module. Therefore, a fire of a battery can be prevented.

Description

Smart battery management system for fire prevention

The present invention relates to a smart battery management system for fire prevention, and more specifically, a smart battery management system capable of preventing fire by synthesizing the SOC (State of Charge) estimation of each battery with improved accuracy and fire prevention information of the collected batteries. It relates to a battery management system.

In systems such as electric vehicles, electric ships, and energy storage devices powered by lithium-ion secondary batteries, BMS plays a role in ensuring the safety and reliability of secondary batteries that supply power to the system. In particular, in order to extend the life of the battery, it is important to constantly maintain the state of charge (SOC) within the normal range (SOC 0% to 100%). If the SOC is too low or high, the battery deteriorates faster than when the SOC is maintained at a medium level. When the battery is over-discharged, components are deteriorated, and when the battery is over-charged, the battery is overheated. In particular, in the case of a fire occurring in an energy storage device, such a battery fire spreads to a large fire, and there is a serious problem in that a large economic loss occurs.

Therefore, accurately estimating the remaining capacity of a battery is essential for increasing battery usage efficiency and lifespan, optimizing performance, and furthermore, preventing fire. Battery usage conditions change greatly from moment to moment, and thus the parameters of the battery also change, so it depends on the degree of battery aging. Accurate estimation techniques are required.

As such a technology, in "Battery state estimation method, apparatus and computer program for performing the method (registration number: 10-2016715)", Ampere-hour Counting Method and Kalman Filter are combined (hybrid type) Disclosed are a method, an apparatus for performing the method, and a computer program for improving the accuracy of battery state estimation by using the Kalman filter method).

As described in the above literature, there is a conventional method of estimating the battery SOC using an ampere-hour counting method. The current integration method is a method of measuring the battery SOC by converting an actual current value into a standard current value using the Peukert equation as shown in Equation 1 below and integrating the value.

Here, SOC ₀ is the initial state of charge of the battery, Q _n is the nominal capacity of the battery, I is the current flowing through the battery in real time over time (t), and a is the current efficiency factor calculated and inferred by the Peukert equation (generally 1 set to).

This current integration method has the advantage that it can be easily implemented, but there is a problem in the accumulation of errors in the estimation of the battery SOC due to the initial value problem and the sensor noise. In other words, accurate sensing of the current sensor is very important because the battery SOC is estimated by the current measured through the current sensor installed in the charging/discharging path of the battery. There is a problem in that a current offset having a difference in the measured value of current occurs, and the error range between the actual battery SOC and the estimated battery SOC increases as time elapses due to the accumulation of the current offset.

A conventional apparatus for estimating a battery state may be modeled as an equivalent circuit by reflecting the electrochemical characteristics of the battery, as shown in FIG. 1 . In FIG. 1, E(t) is an open circuit voltage (OCV), I(t) is a current, R ₁ is an internal resistance of a battery, R ₂ is a polarization resistance, and C is a polarization capacity.

When Kirchhoff's law and current integration method are applied to the equivalent circuit of a battery, the following [Equation 2] to [Equation 5] can be obtained.

where S(t) is the SOC value, Q ₀ is the nominal capacity, and F[S(t)] is a function of SOC and OCV.

In order to solve the above problems, a method using a Kalman filter (KF) algorithm has been studied as shown in FIG. 2 . The Kalman filter (KF) composes a battery equivalent circuit, calculates a system equation that includes the SOC of the battery and the polarization capacity and resistance of the battery equivalent circuit as state variables, and a measurement equation that includes the SOC observed value of the battery as an output variable and estimating the SOC of the battery at a preset cycle by calculating the measured current for the battery, the system equation, and the measurement equation by applying the Kalman filter (KF) algorithm.

Here, the system equation and the measurement equation may be expressed by the following [Equation 6] and [Equation 7], respectively.

Here, S(k+1) and u _c (k+1) are the predicted values of the SOC and the polarization capacity of the battery at the current stage, that is, the SOC and the polarization capacity of the battery, respectively, and S(k) and u _c (k) are SOC and battery polarization capacity of the previous step, respectively. Also, I(k) represents the previous battery current, w ₁ (k) and w ₂ (k) are the noise of the system, v(k) is the voltage measurement noise, and R ₁ , R ₂ and C are the battery equivalents, respectively. These are the circuit's internal resistance, polarization resistance, and polarization capacity.

Meanwhile, in order to improve the accuracy of battery state estimation, a method for predicting battery SOC using a hybrid Kalman filter (HKF) that combines a current integration method and a Kalman filter has been studied. That is, the hybrid Kalman filter (HKF) applies the current integration method by further including the current observation noise to the measured current in the step of calculating the measurement equation among the SOC prediction methods using the Kalman filter (KF), and the following [Equation 8] To calculate a measurement equation such as

Here, z _k is the measurement equation, S(t ₀ ) is the initial SOC value, Q _n is the nominal battery capacity, and H(k) is the current measurement noise.

In order to accurately estimate SOC using such a hybrid Kalman filter (HKF), it is important to reduce errors by substituting accurate values for _{R 2 and C in [Equation 3].}

Conventionally, _{in order to estimate the values of R 2 and C, V 1} and V ₂ are measured based on the starting point of the curve by giving a pulse input during discharging of the battery as shown in FIG. 3, and using this, [Equation 3] There was a method of calculating the values of _{R 2} and C input in ].

However, according to the above-described method, as shown in Fig. 4(a), there is a problem in that the result is significantly different or the error range is large depending on which point is taken as the starting point of the curve. In addition, even though R and C values change according to environmental change factors such as aging and performance degradation due to the accumulation of charge and discharge of the battery, these environmental change factors are not considered, and the initially calculated R ₂ and C values are used. Since the SOC is continuously estimated, there is a limit that the reliability of the SOC prediction may decrease as the number of charge/discharge of the battery increases. Therefore, it is urgently necessary to develop new technology.

KR 10-2016715 B1

Therefore, the present invention is a technique for improving the problems of the prior art, and the accuracy of battery SOC prediction is improved by accurately estimating the R and C values of the battery, which are state variables input to a hybrid Kalman filter (HKF) algorithm. It aims to provide a smart battery management system for a new type of fire prevention that can improve and prevent battery fire by synthesizing the collected fire prevention information.

In order to achieve the above object, a smart battery management system according to an embodiment of the present invention is a smart battery management system (S-BMS) for integrated management of one or more batteries, a hybrid Kalman filter (Hybrid Kalman) a battery state estimation module for estimating a state of charge (SOC) using a filter, HKF), and predicting R and C values of the battery, which are state variables input to the hybrid Kalman filter, using a deep learning technique; a fire prevention information collection module for collecting the cell temperature of each battery, a fire safety check item signal, and a relay contact signal; and a power control module for controlling charging/discharging of a battery by synthesizing signals from the battery state estimation module and the fire prevention information collection module.

The battery state estimating module of the smart battery management system includes: a 1-1 step of measuring R and C values of n sampled batteries from among all batteries; a step 2-1 of calculating _{the SOC n} of each sampling battery through a hybrid Kalman filter algorithm using the measured R and C values as input state variables; _{Step 3-1, comparing the SOC n} of the sampling battery with the ideal SOC to find an optimal approximate value SOC; and step 4-1 of determining the R and C values of the optimal approximate value SOC as R and C values of the entire battery to obtain the calculated optimal R and C values with the R and C measured values of the sampling battery. SOC of each battery that accumulates big data, performs a deep learning algorithm using the big data, and performs a hybrid Kalman filter algorithm using the optimal R and C values and these as input state variables through the deep learning algorithm It is characterized by predicting (Ideal SOC: The optimal value calculated by the battery design value or the optimal value estimated by experiment)

The battery state estimation module of the smart battery management system includes: a 1-2 step of measuring R and C values of one sampling battery sampled from among all batteries; _{SOC m} of a sampling battery is calculated through a hybrid Kalman filter algorithm using _{R m} , C _{m obtained} by adding or subtracting minute increment values ΔR and ΔC to the measured R and C values and the measured R and C values by m as input state variables Step 2-2 to do; _{a step 3-2 of comparing the SOC m} of the sampling battery with the ideal SOC to find an optimal approximate value SOC; and step 4-2 of determining the R and C values of the optimal approximate value SOC as the R and C values of the entire battery to accumulate the optimal R and C values for the sampling battery to obtain big data, It is characterized in that the deep learning algorithm is performed using the big data, and the SOC of each battery subjected to the hybrid Kalman filter algorithm using the optimal R and C values and these as input state variables through the deep learning algorithm is predicted. (Ideal SOC: The optimal value calculated by the battery design value or the optimal value estimated by experiment)

The battery state estimation module of the smart battery management system is characterized in that the battery sampling time is the initial stage of battery installation.

The battery state estimation module of such a smart battery management system performs battery sampling periodically after battery installation to obtain optimal R and C values according to battery usage time as secular big data, and uses the secular change big data It is characterized in that the deep learning algorithm is performed, and the SOC of each battery subjected to the hybrid Kalman filter algorithm using the optimal R and C values that change according to the use of the battery through the deep learning algorithm as an input state variable is predicted.

The fire safety check item signal of this smart battery management system is a power signal, a ground fault detection and interruption (GPT) signal, a power conversion device (PCS) signal, a surge protection device (SPD) signal, an insulation monitoring device (IMD) signal, an overcurrent It is characterized in that it includes any one or more of the blocking (Fuse) signals.

The power control module of such a smart battery management system, when detecting a fire risk in each battery, it is characterized in that the connection of the battery is cut off.

According to the smart battery management system for fire prevention according to the present invention, since the R and C values of the battery, which are input state variables, are predicted through deep learning when the Kalman filter algorithm is applied, the accuracy of estimating the R and C values of the battery as the state variables can be improved, and thus the error range of SOC prediction is reduced, thereby improving the accuracy of SOC prediction.

Accordingly, it is possible to prevent and block problems in which the battery is continuously overcharged or overdischarged due to an SOC prediction error, and overheating due to the accumulation of stress in the battery leads to a fire.

It has the effect of collecting various signals that can cause a fire in the battery and combining it with battery SOC prediction information to prevent battery fire.

In addition, through the smart battery management system according to the present invention, the state of the battery current, voltage, temperature, etc. is monitored in real time to predict the remaining capacity and replacement time to be utilized for system control and to improve battery life through equalization control of battery performance. It has the advantage of reducing maintenance costs by securing system reliability and safety through battery protection and self-diagnosis.

1 is an equivalent circuit diagram reflecting the electrochemical characteristics of a battery.
2 is a flowchart of a conventional Kalman filter (KF) algorithm.
3 is a conventional time-voltage graph when a pulse input is applied during discharging of a battery in order to estimate the values _{of R 2 and C;}
4 is a battery SOC prediction graph using a conventional hybrid Kalman filter (HKF) and a deep learning technique according to an embodiment of the present invention.
5 is a block diagram of a smart battery management system according to an embodiment of the present invention.
6 is a block diagram of a battery SOC prediction algorithm according to an embodiment of the present invention.
7 is a block diagram of a battery SOC prediction algorithm according to another embodiment of the present invention.
8 is an optimal configuration diagram for collecting fire safety check item data according to an embodiment of the present invention.
9 is a block diagram of a smart battery management system according to an embodiment of the present invention.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings 1 to 9. On the other hand, in the drawings and detailed description, drawings and descriptions of configurations and actions that can be easily recognized by those skilled in the art from general BMS, SOC, Kalman filter, etc. are simplified or omitted. In particular, in the drawings and detailed descriptions, detailed descriptions and illustrations of specific technical configurations and actions of elements not directly related to the technical features of the present invention are omitted, and only the technical configurations related to the present invention are briefly illustrated or described. did

The smart battery management system for fire prevention according to an embodiment of the present invention is configured to include a battery state estimation module 100 , a fire prevention information collection module 200 , and a power control module 300 as shown in FIG. 5 .

The battery state estimation module 100 predicts the battery SOC.

According to the above-mentioned "Battery state estimation method, apparatus and computer program for performing the method (registration number: 10-2016715)", as shown in FIG. 4(a), the result greatly depends on which point is taken as the starting point of the curve There was a problem that the difference or error range increased, and environmental change factors such as aging and performance degradation due to the accumulation of charging and discharging of the battery should be considered.

_{Therefore, in the embodiment of the present invention, these V 1} and V ₂ are measured at various points in FIG. 3 using a deep learning technique that considers the number of charging and discharging and the environmental change of the battery, and the values of _{R 2 and C accordingly} By calculating and selecting the most suitable value, it is possible to reduce the error range of SOC prediction as shown in FIG. 4(b).

That is, the battery SOC prediction algorithm according to the embodiment of the present invention may be represented as shown in FIG. 6 . The R and C values of n sampling batteries sampled among all batteries are measured, and the SOC _n of each sampling battery is calculated through a hybrid Kalman filter algorithm using the measured R and C values as input state variables. By comparing the calculated SOC _n with the ideal SOC selected as the optimal value calculated by the battery design value or the optimal value estimated by experimentation, the optimal approximate value SOC is found, and the R and C values of the optimal approximate value SOC are calculated as R and C values of the entire battery. set to the value of C.

In this case, big data may be obtained by accumulating the R and C measured values of the sampling battery and the calculated optimal R and C values. A deep learning algorithm is performed using the acquired big data, and by performing the hybrid Kalman filter algorithm using the optimal R and C values and these as input state variables, each SOC can be predicted for the entire battery to be predicted. will become

By using this battery SOC prediction technique, more precise prediction is possible, and the user's convenience can be promoted because the optimum value of the input state variable is automatically extracted.

A battery SOC prediction algorithm according to another embodiment of the present invention may be represented as shown in FIG. 7 . _{First, SOC 1} is calculated by performing a hybrid Kalman filter algorithm using the first measured R and C values of one sampled battery among all batteries as input state variables.

Next, the values obtained by adding or subtracting minute increment values ΔR and ΔC to the _{measured R 1} , C _{1 values by m orders are set as predicted R m} , C _m values, and a number of predicted R _m , C _m values are used as input state variables. The Kalman filter algorithm is performed to calculate the SOC _{m for each predicted R m} , C _m value, the calculated SOC _m is compared with the ideal SOC to find the best-approximation SOC, and the R, C values of the best-approximation SOC are calculated for the entire battery. It is determined by R and C values.

In this case, big data may be obtained by accumulating optimal R and C values for the sampling battery. A deep learning algorithm is performed using the acquired big data, and by performing the hybrid Kalman filter algorithm using the optimal R and C values and these as input state variables, each SOC can be predicted for the entire battery to be predicted. will become

In particular, the SOC can be predicted more simply than in the case of the above-described embodiment as shown in FIG. 6 by generating a large number of data using one sampled measured data and converting the data into big data.

The battery state estimation module 100 initially predicts the SOC of the battery at the initial stage of battery installation by the above-described method, but enables battery sampling to be periodically and repeatedly performed. This is because the performance of the battery or environmental changes may occur depending on the battery usage time, and since the R and C values are changed accordingly, the optimal SOC value also changes.

Therefore, by periodically repeating battery sampling according to the aging of the battery, it is possible to obtain the optimal R and C values according to the battery usage time as aging big data.

When the deep learning algorithm is performed using the secular big data obtained when predicting each SOC of the entire battery by the method shown in FIG. 6 or 7, the optimal R, If the C value can be calculated and the hybrid Kalman filter algorithm using this value as an input state variable is performed, it is possible to predict the SOC considering the aging of each battery.

The smart battery management system according to an embodiment of the present invention predicts the SOC of the battery in the above-described method through the battery state estimation module 100 , while a fire may occur through the fire prevention information collection module 200 . Detect risk factors in advance.

The fire prevention information collection module 200 is to intensively check factors having a high fire risk in the power system, and collects the temperature of each battery cell, a fire safety check item signal, and a relay contact signal.

Among them, the fire safety inspection item signals are as follows: power signal, ground fault detection and interruption (GPT) signal, power conversion device (PCS) signal, surge protection device (SPD) signal, insulation monitoring device (IMD) signal, overcurrent cutoff (Fuse) It includes any one or more of the signals, and these check items are configured as shown in FIG. 7 to collect signals.

As described above, there is an effect of collecting various signals that can cause a fire in the battery and combining it with battery SOC prediction information to prevent a fire in the battery.

The smart battery management system according to an embodiment of the present invention predicts the SOC of the battery in the battery state estimation module 100 and detects risk factors that may cause a fire in the fire prevention information collection module 200 in advance. On the other hand, when a fire occurs, it is configured to further include a power control module 300 to block in advance so as not to spread.

The power control module 300 controls the charging/discharging of each battery by synthesizing the signals of the battery state estimation module 100 and the fire prevention information collection module 200 . In particular, when the battery detects a fire hazard, it is possible to prevent the spread of fire by blocking the connection of the battery.

In addition, if a monitoring system capable of monitoring the state of battery current, voltage, temperature, etc. in real time through the smart battery management system according to the present invention is further provided, the state of the battery is monitored in real time by an administrator and the remaining capacity of the battery And replacement time can be easily predicted, and this has the advantage of being able to use this for power system control or secure battery lifespan through equalization control of battery performance to secure battery management system reliability and safety and reduce maintenance costs.

In addition, a cloud storage device or a black box may be further provided in order to prevent recurrence by analyzing the cause after a fire occurs. The cloud storage device or the black box includes battery monitoring data received from the battery state estimation module 100 , data for each safety check item received from the fire prevention information collection module 200 , and battery charging/receiving data received from the power control module 300 . By storing the discharge and operation data in the database and making it possible to check and analyze afterward, when a fire occurs, the cause can be analyzed to prevent the recurrence of the fire.

Finally, the smart battery management system according to the present invention including a battery state estimation module 100, a fire prevention information collection module 200, a power control module 300, a monitoring system, a cloud storage device or a black box is shown in FIG. It may be configured as shown.

Therefore, the smart battery management system for fire prevention of the present invention can prevent the accumulation of stress due to continuous overcharging or overdischarging of the battery by accurately predicting the SOC in consideration of the aging change of the battery through the battery state estimation module, It can prevent the battery from overheating. In addition, there is an effect of collecting various signals that can cause a fire in the battery and combining it with battery SOC prediction information to prevent the risk of fire in the battery in advance.

As described above, although the smart battery management system for fire prevention according to an embodiment of the present invention has been illustrated according to the above description and drawings, this is merely an example and within the scope not departing from the technical spirit of the present invention. It will be appreciated by those skilled in the art that various changes and modifications are possible.

100: battery state estimation module
200: fire prevention information collection module
300: power control module

Claims (7)

In a smart battery management system (S-BMS) for integrated management of one or more batteries,
A battery that estimates the state of charge (SOC) using a hybrid Kalman filter (HKF), but predicts R and C values of the battery, which are state variables input to the hybrid Kalman filter, using a deep learning technique state estimation module;
a fire prevention information collection module for collecting the cell temperature of each battery, a fire safety check item signal, and a relay contact signal; and
A smart battery management system for fire prevention, characterized in that it comprises a; power control module for controlling the charging/discharging of the battery by synthesizing the signals of the battery state estimation module and the fire prevention information collection module. According to claim 1,
The battery state estimation module,
Step 1-1 of measuring R and C values of n sampling batteries sampled from among all batteries;
a step 2-1 of calculating _{the SOC n} of each sampling battery through a hybrid Kalman filter algorithm using the measured R and C values as input state variables;
_{Step 3-1, comparing the SOC n} of the sampling battery with the ideal SOC to find an optimal approximate value SOC; and
By performing the algorithm of step 4-1; determining the R and C values of the optimal approximate value SOC as the R and C values of the entire battery.
Acquiring big data by accumulating the R, C measured values of the sampling battery and the calculated optimal R, C values,
Fire, characterized in that by performing a deep learning algorithm using the big data, predicting the SOC of each battery that has performed the hybrid Kalman filter algorithm using the optimal R and C values and these as input state variables through the deep learning algorithm Smart battery management system for prevention.
(Ideal SOC: The optimal value calculated by the battery design value or the optimal value estimated by experiment) According to claim 1,
The battery state estimation module,
a 1-2 step of measuring R and C values of one sampling battery sampled from among all batteries;
_{SOC m} of a sampling battery is calculated through a hybrid Kalman filter algorithm using _{R m} , C _{m obtained} by adding or subtracting minute increment values ΔR and ΔC to the measured R and C values and the measured R and C values by m as input state variables Step 2-2 to do;
_{a step 3-2 of comparing the SOC m} of the sampling battery with the ideal SOC to find an optimal approximate value SOC; and
By performing the algorithm of step 4-2; determining the R and C values of the optimal approximate value SOC as the R and C values of the entire battery.
Acquiring big data by accumulating optimal R and C values for the sampling battery,
Fire, characterized in that by performing a deep learning algorithm using the big data, predicting the SOC of each battery that has performed the hybrid Kalman filter algorithm using the optimal R and C values and these as input state variables through the deep learning algorithm Smart battery management system for prevention.
(Ideal SOC: The optimal value calculated by the battery design value or the optimal value estimated by experiment) 4. The method of claim 2 or 3,
The battery state estimation module,
A smart battery management system for fire prevention, characterized in that the battery sampling time is at the initial stage of battery installation. 5. The method of claim 4,
The battery state estimation module,
Battery sampling is repeated periodically after battery installation.
Acquire the optimal R and C values according to the battery usage time as secular big data,
SOC of each battery that performs a deep learning algorithm using the aging big data Smart battery management system for fire prevention, characterized in that predicting. According to claim 1,
The fire safety check item signal is,
Power signal, ground fault detection and cutoff (GPT) signal, power conversion device (PCS) signal, surge protection device (SPD) signal, insulation monitoring device (IMD) signal, characterized in that it includes any one or more of the overcurrent cutoff (Fuse) signal Smart battery management system for fire prevention. According to claim 1,
The power control module,
A smart battery management system for fire prevention, characterized in that when each battery detects a fire risk, the connection to the corresponding battery is cut off.

Images