Data anomaly detection method based on battery behavior characteristics

Abstract

The invention relates to a data anomaly detection method based on battery behavior characteristics. The method comprises the following steps: S1, acquiring voltage, temperature and current of each single battery in a battery system BatS1; s2, judging whether the temperature and the current of the battery in the battery system BatS1 from the first moment to the current moment are in a steady state or not; s3, in the battery system BatS1, from the first moment to the current moment, when all battery temperatures and currents are in a steady state, starting from the current moment, carrying out data anomaly detection on battery voltages; and S4, eliminating batteries with poor battery parameter consistency in the battery system BatS1 to form a battery system BatS2, and in the battery system BatS2, re-recording the battery voltage, temperature and current data from the first moment to the current moment, and carrying out data anomaly detection on the battery voltage. Based on voltage, temperature, current and other data acquired by a battery management system, data anomaly detection is realized by combining battery behavior characteristics in charging and discharging states, and powerful support is provided for ensuring safe and stable operation of the battery.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery management, and in particular relates to a data anomaly detection method based on battery behavior characteristics. Background Art

[0002] In the modern world driven by electricity, the battery management system plays a vital role. It is not only responsible for maintaining the stable operation of the battery cells, but also bears the heavy task of monitoring the battery status and preventing overcharging and over-discharging, which is not only to improve the battery life, but also to ensure safety. Therefore, accurate and intelligent battery data anomaly detection methods are particularly critical.

[0003] The battery management system needs to continuously collect key data such as battery voltage, temperature, and current. These data are like a precise ruler that can accurately reflect the health of the battery. However, when these data are abnormal, whether due to hardware aging, software errors, or other unknown factors, they may have a negative impact on battery performance and even cause safety issues.

[0004] At present, although many methods have been proposed to detect anomalies in battery data, these methods often have problems such as low efficiency, high false alarm rate or difficulty in real-time detection. Although there have been breakthroughs in predicting the life and performance of lithium-ion batteries at home and abroad, research on its abnormal point detection and energy storage system maintenance is still insufficient.

[0005] As we all know, the K-nearest neighbor method is a commonly used method for outlier detection. However, in practical applications, due to the large number of lithium battery test samples, each test sample needs to calculate the sample distance from itself to the remaining samples. This process is computationally intensive, resulting in high algorithm complexity and low computational efficiency. To solve the problem of the huge amount of computation required by the K-nearest neighbor method in lithium battery outlier detection, it is necessary to comprehensively consider multiple aspects such as algorithm optimization, data dimensionality reduction, parallel computing, improved algorithm models, and intelligent optimization strategies in order to achieve the purpose of improving computational efficiency and accuracy. However, the application process is extremely complex, the number of calculations is complicated, the operation consumption is large, and a long response time is required, which is not convenient for engineering applications. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a data anomaly detection method based on battery behavior characteristics. The method is easy to implement, can quickly realize battery data anomaly detection, effectively improve the accuracy and efficiency of battery data anomaly detection, and provide strong support for ensuring the safety and stable operation of the battery.

[0007] The present invention provides a data anomaly detection method based on battery behavior characteristics, comprising the following steps: S1: Obtain the voltage, temperature and current of each single cell in the battery system BatS1; S2: Determine whether the battery temperature and current of the battery in the battery system BatS1 are in a steady state from the first moment to the current moment; S3: In the battery system BatS1, from the first moment to the current moment, all battery temperatures and currents are in a steady state. From the current moment, the battery voltage is tested for data anomalies; S4: In the battery system BatS1, batteries with poor battery parameter consistency are eliminated to form a battery system BatS2. In the battery system BatS2, the battery voltage, temperature, and current data from the first moment to the current moment are re-recorded, and data anomaly detection is performed on the battery voltage.

[0008] Preferably, the step S1 includes recording the voltage data V of the i-th battery in the battery system BatS1 at the j-th moment. ij , Temperature data T ij , and the current data I at the jth moment j .

[0009] Preferably, in step S2, when the absolute difference of temperature data of the i-th battery at two adjacent moments is controllable, and the temperature data at two adjacent moments are both in the temperature controllable interval [Tdown, Tup], the temperature of the i-th battery is in a steady state; when the absolute difference of current data at two adjacent moments is controllable, and the current data at two adjacent moments are in the current controllable interval [Idown, Iup], the current of the i-th battery is in a steady state.

[0010] Preferably, step S3 comprises the following steps: S31: Starting from the current time n, determine whether the voltage of each battery in the battery system is within the controllable range [Vdown, Vup] and whether it is consistent with the average voltage of the battery system at the current time n. The deviation does not exceed the set threshold , record the voltage in the battery system in the controllable range [Vdown, Vup] and the average voltage of the battery system The number A of deviations not exceeding the set threshold Vavg_Th; The average voltage of the battery system at the current time n The calculation formula is as follows: , Among them, m is the number of batteries, and n is the current moment; Threshold The calculation formula is as follows: , , , in, is the cumulative calculation coefficient of the battery temperature at the current moment of the i-th section, is the current accumulation calculation coefficient at the current moment; Ce is the rated capacity of the battery in the battery system; It is the reference voltage threshold of 1C current rate at normal battery temperature of 25 degrees Celsius, which is equal to Ce in value; S32: From the current moment on, determine whether the overall voltage of the battery system has reached a steady state, set a threshold value A_Th, whose value is greater than or equal to 0 and less than or equal to 1; when A / m>A_Th, the overall voltage of the battery has reached a steady state; S33: Starting from the current moment, record the number of times B that the voltage deviation value of the i-th battery in the battery system at the current moment and the voltage at the previous moment continuously exceeds the set threshold value Vi_adj_th; set the threshold value B_Th, if B is greater than B_Th, and except for the i-th battery, there is no situation where B is greater than B_Th for other batteries, then it is determined that all battery temperatures and all battery currents are in a steady state, and the voltage data detection of the i-th battery is abnormal.

[0011] Preferably, step S4 includes the following steps: S41: From the current moment on, in the battery system, each battery will be charged and discharged individually, and the battery will be charged to a cut-off voltage or discharged to a cut-off voltage, and batteries with poor battery parameter consistency will be eliminated, and batteries with good battery parameter consistency will be selected; S42: In the battery system, the batteries with good battery parameter consistency form a new battery system BatS2. After charging to the battery cut-off voltage or discharging to the battery cut-off voltage, the battery voltage, temperature, and current data from the first moment to the current moment are re-recorded, and the battery system BatS2 is checked to check whether the battery temperature is in the controllable range [Tdown, Tup], whether the current is in the controllable range [Idown, Iup], and whether the voltage is in the controllable range [Vdown, Vup] at each moment, and detect batteries with abnormal battery voltage data.

[0012] Preferably, selecting batteries with good battery parameter consistency in step S41 includes the following steps: S411: Place the battery system in temperature; use constant current I_low to charge to the cut-off voltage, then rest for h hours, and repeat several times; use constant current I_up to discharge to the cut-off voltage, then rest for h hours, and repeat several times; during this period, record the temperature, voltage, current and cumulative capacity of each battery at intervals of 1 second, where the cumulative capacity of the battery is the continuous cumulative value of the product of the absolute value of the current and the time; S412: Select the batteries whose battery voltage, battery temperature, battery current, and battery cumulative capacity at each moment are less than the set threshold Th1 from the average voltage, battery temperature, battery current, and battery cumulative capacity of the battery system BatS1; S413: Select the battery whose maximum and minimum differences of battery voltage, battery temperature, battery current, and battery cumulative capacity at each moment do not exceed a set threshold value Th2; S414: Select the battery whose maximum and minimum differences of the battery cumulative capacity, charge / discharge and open circuit cumulative time do not exceed the set threshold Th3.

[0013] Preferably, the step S42 includes the following steps: S421: During the period from the first moment to the current moment, the batteries in the battery system BatS2 are charged and discharged using the same current, and the voltage and temperature of each battery in the battery system BatS2 are summed and averaged to obtain the accumulated temperature value ST(1-k) of the dth battery from the first moment to the kth moment. d , the average value of the accumulated temperature of battery No. d from moment 1 to moment k is STavg(1-k) d , the accumulated value of the voltage of battery No. d from the 1st moment to the kth moment SV(1-k) d , the average value of the accumulated voltage of battery No. d from moment 1 to moment k SVavg(1-k) d ; Calculate the battery SOC using the battery rated capacity and the ampere-hour integration method at the same time; S422: Divide the battery from the first moment to the current moment into multiple detection moments according to the same moment q. Test time period; and make the following judgments: In the same test period, the difference between the maximum and minimum SOC values ​​of different batteries is less than SOC_Th2. The difference between the maximum and minimum average values ​​of the accumulated values ​​of the battery voltages of different batteries during the same detection period is less than SVavg_Th; If all of the above conditions are met, proceed to step S423, otherwise, if one of them is not met, do not proceed to step S423; S423: Select batteries with high pre-voltage C and batteries with low pre-voltage C from the battery system BatS2, and after removing the batteries with high pre-voltage C and batteries with low pre-voltage C from the battery system BatS2, calculate the average value SVavg_R of the remaining batteries SVavg, and calculate the average value SV _R of the remaining batteries SV; set a threshold SVavg_Th for the batteries with high pre-voltage C and batteries with low pre-voltage C, if the absolute value of the difference between its SVavg and SVavg_R is less than SVavg_Th, the battery voltage is normal, otherwise the battery voltage data detection is abnormal.

[0014] The deviation of the battery SOC difference in two adjacent detection periods of the same battery is less than SOC_Th1, otherwise the battery voltage data detection is abnormal; the slope deviation of the battery temperature accumulated value in two adjacent detection periods of the same battery is less than ST_K, otherwise the battery voltage data detection is abnormal; the slope deviation of the battery voltage accumulated value in two adjacent detection periods of the same battery is less than SV_K, otherwise the battery voltage data detection is abnormal.

[0015] The present invention has the following technical effects: Based on the voltage, temperature, current and other data collected by the battery management system, combined with the battery behavior characteristics under charging and discharging conditions, data anomaly detection is achieved. The method is easy to implement and can effectively improve the accuracy and efficiency of battery data anomaly detection, providing strong support for ensuring the safety and stable operation of the battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of a data anomaly detection method based on battery behavior characteristics of the present invention; Figure 2 The present invention divides the battery time into multiple detection times according to the same time q. Test Time period diagram. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings.

[0018] like Figure 1 As shown, a data anomaly detection method based on battery behavior characteristics includes the following steps: S1: Obtain the voltage, temperature and current of each single cell in the battery system BatS1; Record the voltage data V of the i-th battery at the j-th moment in the battery system BatS1 ij , Temperature data T ij , and the current data I at the jth moment j .

[0019] S2: Determine whether the battery temperature and current of the battery in the battery system BatS1 are in a steady state from the first moment to the current moment; When the absolute difference of the temperature data of the i-th battery at two adjacent moments is controllable, and the temperature data at two adjacent moments are both in the temperature controllable interval [Tdown, Tup], the temperature of the i-th battery is in a steady state; when the absolute difference of the current data at two adjacent moments is controllable, and the current data at two adjacent moments are in the current controllable interval [Idown, Iup], the current of the i-th battery is in a steady state.

[0020] When [Tdown, Tup] is [0, 45], |T 11 -T 12 |<0.5℃ and T 11 、T 12 ∈[0, 45], the temperature of the battery in Section 1 is in a steady state. When [Idown, Iup] is [50, 100], |I1-I2|<0.5 ampere and I1, I2∈[50, 100], the current is in a steady state.

[0021] S3: In the battery system BatS1, from the first moment to the current moment, all battery temperatures and currents are in a steady state. From the current moment, the battery voltage is tested for data anomalies; S31: Starting from the current time n, determine whether the voltage of each battery in the battery system is within the controllable range [Vdown, Vup] and whether it is consistent with the average voltage of the battery system at the current time n. The deviation does not exceed the set threshold , record the voltage in the battery system in the controllable range [Vdown, Vup] and the average voltage of the battery system The number A of deviations not exceeding the set threshold Vavg_Th; Average voltage of the battery system at the current time n The calculation formula is as follows: , Wherein, m is the number of batteries, and n is the current moment (i.e., the last current moment among the 1st moment, the 2nd moment, ... the jth moment, ..., the nth moment. n is a variable value, for example, if the current moment is the 6th moment, n is 6; if the previous moment, i.e., the 5th moment, is the current moment of the previous moment, n of the previous moment is 5);

[0022] Threshold The calculation formula is as follows: , , , in, is the cumulative calculation coefficient of the battery temperature at the current moment of the i-th section, is the current accumulation calculation coefficient at the current moment; Ce is the rated capacity of the battery in the battery system; It is the reference voltage threshold of 1C current rate at normal battery temperature of 25 degrees Celsius, which is equal to Ce in value; Take Ce as 100Ah and three lithium iron phosphate batteries connected in series to form a battery system. The negative current is specified as the charging current value, and the positive current is specified as the discharging current value. The voltage in the battery system is in the controllable range [Vdown, Vup] of [3000, 3450]. The parameter calculation table is shown in Table 1:

[0023] Table 1 Parameter calculation table (three batteries) Time(s) First moment (1s) Second moment (10s) The third moment (20s) I(A) 70.01 70.00 69.98 <![CDATA[V1(mv)]]> 3320 3319 3320 <![CDATA[V2(mv)]]> 3322 3321 3322 <![CDATA[V3(mv)]]> 3323 3322 3323 Vtavg(mv) 3322 3321 3322 <![CDATA[T1(℃)]]> 32 32 33 <![CDATA[T2(℃)]]> 28 28 29 <![CDATA[T3(℃)]]> 25 25 26 KI 1.30 1.30 1.30 <![CDATA[KT1]]> 0.72 0.72 0.70 <![CDATA[KT2]]> 0.88 0.88 0.86 <![CDATA[KT3]]> 1.00 1.00 0.98 <![CDATA[Vavg_Th1]]> 93.60 93.60 91.00 <![CDATA[Vavg_Th2]]> 114.40 114.40 111.80 <![CDATA[Vavg_Th3]]> 130.00 130.00 127.40 The temperature array of the first battery at the first moment is T1 =

[32] . At this time, the current moment is the first moment. Substitute Formula to obtain KT 11 is 0.72; The temperature array from the first moment to the second moment is T1=[32,32]. At this time, the current moment is the second moment. Substitute Formula to obtain KT 12 is 0.72; The temperature array from the 1st moment to the 3rd moment is T1=[32,32,33]. The current moment is the 3rd moment. Substitute Formula to obtain KT 13 0.70 At each moment, the voltages of the three batteries are within the controllable range of the battery system voltage, and the deviation from the average voltage does not exceed the set threshold, and the statistic A is 3.

[0024] S32: From the current moment on, determine whether the overall voltage of the battery system has reached a steady state, set a threshold value A_Th, whose value is greater than or equal to 0 and less than or equal to 1; when A / m>A_Th, the overall voltage of the battery has reached a steady state; S33: Starting from the current moment, record the number of times B that the voltage deviation value of the i-th battery in the battery system at the current moment and the voltage at the previous moment continuously exceeds the set threshold value Vi_adj_th; set the threshold value B_Th, if B is greater than B_Th, and except for the i-th battery, there is no situation where B is greater than B_Th for other batteries, then it is determined that all battery temperatures and all battery currents are in a steady state, and the voltage data detection of the i-th battery is abnormal.

[0025] Take Ce as 100Ah and three lithium iron phosphate batteries (the maximum value of i is 3) are connected in series to form a battery system. The current direction is specified as negative for charging and positive for discharging. is 50mv. The voltage in the battery system is in the controllable range [Vdown, Vup] of [3000, 3450]. Vi_adj_th is set to 10mv, B_Th value is set to 3, and AN_Th value is set to 0.5. The battery system parameter calculation table is shown in Table 2:

[0026] Table 2 Battery system parameter calculation table (three batteries) Time(s) First moment (1s) Second moment (10s) The third moment (20s) Moment 4 (30s) Moment 5 (40s) Moment 6 (50s) I(A) 55 55 55 55 55 55 <![CDATA[V1(mv)]]> 3120 3132 3145 3125 3155 3125 <![CDATA[V2(mv)]]> 3292 3292 3292 3292 3292 3292 <![CDATA[V3(mv)]]> 3293 3293 3293 3293 3293 3293 Vtavg(mv) 3235 3239 3243 3236 3246 3236 <![CDATA[T1(℃)]]> 27 27 27 27 27 27 <![CDATA[T2(℃)]]> 28 28 28 28 28 28 <![CDATA[T3(℃)]]> 26 26 26 26 26 26 KI 1.5 1.5 1.5 1.5 1.5 1.5 <![CDATA[KT1]]> 0.92 0.92 0.92 0.92 0.92 0.92 <![CDATA[KT2]]> 0.88 0.88 0.88 0.88 0.88 0.88 <![CDATA[KT3]]> 0.96 0.96 0.96 0.96 0.96 0.96 <![CDATA[Vavg_Th1]]> 138 138 138 138 138 138 <![CDATA[Vavg_Th2]]> 132 132 132 132 132 132 <![CDATA[Vavg_Th3]]> 144 144 144 144 144 144 The current time is the 3rd time. From the 1st time to the 3rd time, the first battery, |T 11 -T 12 |<0.5℃,|T 12 -T 13 |<0.5℃ and T 11 , T 12 , T 13 ∈[0, 45], the temperature of the first battery is in a steady state. Similarly, the temperature of the second and third batteries is in a steady state. From the first moment to the third moment, |I1-I2|<0.5A, |I2-I3|<0.5A and I1, I2, I3∈[50, 100], the current is in a steady state.

[0027] Starting from the current moment, the third moment, the voltages of the three batteries in the battery system are all in the controllable range [3000,3450]. At the third moment, the deviation value between the three batteries and the average voltage Vtavg of the battery system is [98,49,50], which is less than the variable Vavg_th=[138,132,144]. At the third moment, A / i (3 / 3)>Ai_Th (0.5), and the overall voltage of the battery system reaches a steady state. At the fourth moment, the deviation value between the three batteries and the average voltage Vtavg of the battery system is [111,56,57], which is less than the variable Vavg_th=[138,132,144]. At the fourth moment, A / i (3 / 3)>Ai_ Th (0.5), the overall voltage of the battery system reaches a steady state; at the 5th moment, the deviation value between the three batteries and the average voltage Vtavg of the battery system is [91, 46, 47], which is less than the variable Vavg_th = [138, 132, 144]. At the 5th moment, A / i (3 / 3)>Ai_Th (0.5), and the overall voltage of the battery system reaches a steady state; at the 6th moment, the deviation value between the three batteries and the average voltage Vtavg of the battery system is [111, 56, 57], which is less than the variable Vavg_th = [138, 132, 144]. At the 6th moment, A / i (3 / 3)>Ai_Th (0.5), and the overall voltage of the battery system reaches a steady state.

[0028] From the 3rd moment to the 6th moment, the number of times B that the voltage deviation value of the first battery of the battery system continuously exceeds the set threshold value Vi_adj_th (10mv) at two adjacent moments is recorded as 4 times, and the B_Th value is set to 3 times. It is determined that all battery temperatures and all battery currents are in a steady state, and the voltage data detection of the first battery is abnormal.

[0029] S4: In the battery system BatS1, batteries with poor battery parameter consistency are eliminated to form a battery system BatS2. In the battery system BatS2, the battery voltage, temperature, and current data from the first moment to the current moment are re-recorded, and data anomaly detection is performed on the battery voltage.

[0030] S41: From the current moment on, in the battery system, each battery will be charged and discharged individually, and the battery will be charged to a cut-off voltage or discharged to a cut-off voltage, and batteries with poor battery parameter consistency will be eliminated, and batteries with good battery parameter consistency will be selected; The step S41 of selecting batteries with good battery parameter consistency includes the following steps: S411: Place the battery system in temperature; use constant current I_low (0.1C) to charge to the cut-off voltage, then let it rest for 2 hours, and repeat for 3 times; use constant current I_up (1C) to discharge to the cut-off voltage, let it rest for 2 hours, and repeat for 2 times; during this period, record the temperature, voltage, current and cumulative capacity of each battery at 1 second intervals, where the cumulative capacity is the continuous accumulation of the product of the absolute value of the current and the time; S412: Select the batteries whose battery voltage, battery temperature, battery current, and battery cumulative capacity at each moment are less than the set threshold Th1 from the average voltage, battery temperature, battery current, and battery cumulative capacity of the battery system BatS1; S413: Select the battery whose maximum and minimum differences of battery voltage, battery temperature, battery current, and battery cumulative capacity at each moment do not exceed a set threshold value Th2; S414: Select the battery whose maximum and minimum differences of the battery cumulative capacity, charge / discharge and open circuit cumulative time do not exceed the set threshold Th3.

[0031] A battery system consisting of three lithium iron phosphate batteries with a Ce value of 100Ah connected in series is selected, and the battery parameter table in the battery system is shown in Table 3: time The X1 Moment Moment X2 Moment X3 <![CDATA[V1(mv)]]> 3338 2998 3650 <![CDATA[V2(mv)]]> 3345 2992 3650 <![CDATA[V3(mv)]]> 3340 2995 3650 Voltage average value (mv) 3341 2995 3650 <![CDATA[T1(℃)]]> 27 26 32 <![CDATA[T2(℃)]]> 26 25 33 <![CDATA[T3(℃)]]> 25 24 32 Temperature average(℃) 26 25 31 <![CDATA[I1(A)]]> 10 10 10 <![CDATA[I2(A)]]> 10 10 10 <![CDATA[I3(A)]]> 10 10 10 Average current(A) 10 10 10 1# cumulative capacity (Ah) 0 100.025 200.025 2# cumulative capacity (Ah) 0 100.039 200.050 3# Cumulative capacity (Ah) 0 100.031 200.037 Cumulative average capacity (Ah) 0 100.032 200.037 1# Cumulative time (s) 0 43210 79211 2# Cumulative time (s) 0 43215 79220 3# Cumulative time (s) 0 43212 79215 Set the threshold Th1=[10,2,2,2], all are within the set threshold Th1.

[0032] Set threshold Th2=[10,2,2,2], all are within the set threshold Th2.

[0033] Set the threshold Th3=[2,20], all are within the set threshold Th3.

[0034] After screening by thresholds Th1, Th2, and Th3, batteries with poor battery parameter consistency are eliminated. The first battery, the second battery, and the third battery in Table 3 have good consistency.

[0035] S42: In the battery system, the batteries with good battery parameter consistency form a new battery system BatS2. After charging to the battery cut-off voltage or discharging to the battery cut-off voltage, the battery voltage, temperature, and current data from the first moment to the current moment are re-recorded, and the battery system BatS2 is checked to check whether the battery temperature is in the controllable range [Tdown, Tup], whether the current is in the controllable range [Idown, Iup], and whether the voltage is in the controllable range [Vdown, Vup] at each moment, and detect batteries with abnormal battery voltage data.

[0036] S421: During the period from the first moment to the current moment, the batteries in the battery system BatS2 are charged and discharged using the same current, and the voltage and temperature of each battery in the battery system BatS2 are summed and averaged, such as Figure 2 As shown, the accumulated value ST(1-k) of the temperature of the dth battery from the 1st moment to the kth moment is obtained. d , the average value of the accumulated temperature of battery No. d from moment 1 to moment k is STavg(1-k) d , the accumulated value of the voltage of battery No. d from the 1st moment to the kth moment SV(1-k) d , the average value of the accumulated voltage of battery No. d from moment 1 to moment k SVavg(1-k) d ; Calculate the battery SOC using the battery rated capacity and the ampere-hour integration method at the same time; S422: Divide the battery from the first moment to the current moment into multiple detection moments according to the same moment q. Test time period; and make the following judgments: In the same test period, the difference between the maximum and minimum SOC values ​​of different batteries is less than SOC_Th2. The difference between the maximum and minimum average values ​​of the accumulated values ​​of the battery voltages of different batteries during the same detection period is less than SVavg_Th; If all of the above conditions are met, proceed to step S423, otherwise, if one of them is not met, do not proceed to step S423; Ce is a battery system composed of three 100Ah lithium iron phosphate batteries connected in series. In the first period, the SOC(q)1 of the first battery is the maximum value, the SOC(q)3 of the third battery is the minimum value, and the difference between the maximum and minimum values ​​is less than SOC_Th2; In the first period, the first battery ST(1-q)1 is the maximum value, the third battery ST(1-q)3 is the minimum value, and the difference between the maximum and minimum values ​​is less than STavg_Th; If the above conditions are met, proceed to step S423.

[0037] S423: Select the battery with high C before voltage and the battery with low C before voltage from the battery system BatS2, where C is the number of the batteries taken. After removing the battery with high C before voltage and the battery with low C before voltage from the battery system BatS2, calculate the average value SVavg_R of the remaining batteries SVavg and calculate the average value SV_R of the remaining batteries SV; set the threshold SVavg_Th for the battery with high C before voltage and the battery with low C before voltage, if the absolute value of the difference between its SVavg and SVavg_R is less than SVavg_Th, the battery voltage is normal, otherwise the battery voltage data detection is abnormal.

[0038] The deviation of the battery SOC difference in two adjacent detection periods of the same battery is less than SOC_Th1, otherwise the battery voltage data detection is abnormal; the slope deviation of the battery temperature accumulated value in two adjacent detection periods of the same battery is less than ST_K, otherwise the battery voltage data detection is abnormal; the slope deviation of the battery voltage accumulated value in two adjacent detection periods of the same battery is less than SV_K, otherwise the battery voltage data detection is abnormal.

[0039] In a battery system composed of three lithium iron phosphate batteries with a Ce of 100Ah connected in series, in the first period of the battery system BatS2, if at the qth moment, the first battery is the highest and the third battery is the lowest, calculate the average SVavg(1-q) of all the remaining batteries SVavg_R(1-q) The absolute value of the difference between SVavg(1-q)1 and SVavg_R(1-q) of the first battery is less than SVavg_Th, which is normal; The absolute value of the difference between SVavg(1-q)5 and SVavg_R(1-q) of the third battery is greater than or equal to SVavg_Th, which means it is an abnormal battery.

[0040] The deviation of the difference between SOC(q)1 at time q in the first period of battery No. 1 and SOC(q)1 at time q in the second period of battery No. 1 is less than SOC_Th1, which is normal; the deviation of the slope between ST(1-q)1 in the first period of battery No. 1 and ST(1-2q)1 in the second period of battery No. 1 is less than ST_K, which is normal; the deviation of the slope between SV(1-q)1 in the first period of battery No. 1 and SV(1-2q)1 in the second period of battery No. 1 is less than SV_K, which is normal.

[0041] The embodiments described above are only descriptions of the preferred modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A data anomaly detection method based on battery behavior characteristics, characterized in that: The steps include: S1: Obtain the voltage, temperature and current of each single cell in the battery system BatS1; S2: Determine whether the battery temperature and current of the battery in the battery system BatS1 are in a steady state from the first moment to the current moment; S3: In the battery system BatS1, from the first moment to the current moment, all battery temperatures and currents are in a steady state. From the current moment, the battery voltage is tested for data anomalies; S4: In the battery system BatS1, batteries with poor battery parameter consistency are eliminated to form a battery system BatS2. In the battery system BatS2, the battery voltage, temperature, and current data from the first moment to the current moment are re-recorded, and data anomaly detection is performed on the battery voltage.

2. The data anomaly detection method based on battery behavior characteristics according to claim 1 is characterized in that: The step S1 includes recording the voltage data V of the i-th battery at the j-th moment in the battery system BatS1. ij , Temperature data T ij , and the current data I at the jth moment j .

3. The data anomaly detection method based on battery behavior characteristics according to claim 2 is characterized in that: In step S2, when the absolute difference of the temperature data of the i-th battery at two adjacent moments is controllable, and the temperature data at two adjacent moments are both in the temperature controllable interval [Tdown, Tup], the temperature of the i-th battery is in a steady state; when the absolute difference of the current data at two adjacent moments is controllable, and the current data at two adjacent moments are in the current controllable interval [Idown, Iup], the current of the i-th battery is in a steady state.

4. The data anomaly detection method based on battery behavior characteristics according to claim 3 is characterized in that: The step S3 comprises the following steps: S31: Starting from the current time n, determine whether the voltage of each battery in the battery system is within the controllable range [Vdown, Vup] and whether it is consistent with the average voltage of the battery system at the current time n. The deviation does not exceed the set threshold , record the voltage in the battery system in the controllable range [Vdown, Vup] and the average voltage of the battery system The number A of deviations not exceeding the set threshold Vavg_Th; The average voltage of the battery system at the current time n The calculation formula is as follows: ， Among them, m is the number of batteries, and n is the current moment; Threshold The calculation formula is as follows: ， ， ， in, is the cumulative calculation coefficient of the battery temperature at the current moment of the i-th section, is the current accumulation calculation coefficient at the current moment; Ce is the rated capacity of the battery in the battery system; It is the reference voltage threshold of 1C current rate at normal battery temperature of 25 degrees Celsius, which is equal to Ce in value; S32: From the current moment on, determine whether the overall voltage of the battery system has reached a steady state, set a threshold value A_Th, whose value is greater than or equal to 0 and less than or equal to 1; when A / m>A_Th, the overall voltage of the battery has reached a steady state; S33: Starting from the current moment, record the number of times B that the voltage deviation value of the i-th battery in the battery system at the current moment and the voltage at the previous moment continuously exceeds the set threshold value Vi_adj_th; set the threshold value B_Th, if B is greater than B_Th, and except for the i-th battery, there is no situation where B is greater than B_Th for other batteries, then it is determined that all battery temperatures and all battery currents are in a steady state, and the voltage data detection of the i-th battery is abnormal.

5. The data anomaly detection method based on battery behavior characteristics according to claim 1 is characterized in that: The step S4 includes the following steps: S41: From the current moment on, in the battery system, each battery will be charged and discharged individually, and the battery will be charged to a cut-off voltage or discharged to a cut-off voltage, and batteries with poor battery parameter consistency will be eliminated, and batteries with good battery parameter consistency will be selected; S42: In the battery system, the batteries with good battery parameter consistency form a new battery system BatS2, and after charging to the battery cut-off voltage or discharging to the battery cut-off voltage, re-record the battery voltage, temperature, and current data from the first moment to the current moment, check in the battery system BatS2, at each moment whether the battery temperature is in the controllable range [Tdown, Tup], whether the current is in the controllable range [Idown, Iup], whether the voltage is in the controllable range [Vdown, Vup], and detect batteries with abnormal battery voltage data.

6. The data anomaly detection method based on battery behavior characteristics according to claim 5 is characterized in that: The step S41 of selecting batteries with good battery parameter consistency includes the following steps: S411: Place the battery system in temperature; use constant current I_low to charge to the cut-off voltage, then rest for h hours, and repeat several times; use constant current I_up to discharge to the cut-off voltage, then rest for h hours, and repeat several times; during this period, record the temperature, voltage, current and cumulative capacity of each battery at intervals of 1 second, where the cumulative capacity of the battery is the continuous cumulative value of the product of the absolute value of the current and the time; S412: Select the batteries whose battery voltage, battery temperature, battery current, and battery cumulative capacity at each moment are less than the set threshold Th1 from the average voltage, battery temperature, battery current, and battery cumulative capacity of the battery system BatS1; S413: Select the battery whose maximum and minimum differences of battery voltage, battery temperature, battery current, and battery cumulative capacity at each moment do not exceed a set threshold value Th2; S414: Select the battery whose maximum and minimum differences of the battery cumulative capacity, charge / discharge and open circuit cumulative time do not exceed the set threshold Th3.

7. The data anomaly detection method based on battery behavior characteristics according to claim 5 is characterized in that: The step S42 includes the following steps: S421: During the period from the first moment to the current moment, the batteries in the battery system BatS2 are charged and discharged using the same current, and the voltage and temperature of each battery in the battery system BatS2 are summed and averaged to obtain the accumulated temperature value ST(1-k) of the dth battery from the first moment to the kth moment. d , the average value of the accumulated temperature of battery No. d from moment 1 to moment k is STavg(1-k) d , the accumulated value of the voltage of battery No. d from the 1st moment to the kth moment SV(1-k) d , the average value of the accumulated voltage of battery No. d from moment 1 to moment k SVavg(1-k) d ; Calculate the battery SOC using the battery rated capacity and the ampere-hour integration method at the same time; S422: Divide the battery from the first moment to the current moment into multiple detection moments according to the same moment q. Test time period; and make the following judgments: In the same test period, the difference between the maximum and minimum SOC values ​​of different batteries is less than SOC_Th2. The difference between the maximum and minimum average values ​​of the accumulated values ​​of the battery voltages of different batteries during the same detection period is less than SVavg_Th; If all of the above conditions are met, proceed to step S423, otherwise, if one of them is not met, do not proceed to step S423; S423: select the battery with high pre-voltage C and the battery with low pre-voltage C from the battery system BatS2, and after removing the battery with high pre-voltage C and the battery with low pre-voltage C from the battery system BatS2, calculate the average value SVavg_R of the remaining battery SVavg, and calculate the average value SV_R of the remaining battery SV; set the threshold SVavg_Th for the battery with high pre-voltage C and the battery with low pre-voltage C, if the absolute value of the difference between its SVavg and SVavg_R is less than SVavg_Th, the battery voltage is normal, otherwise the battery voltage data detection is abnormal; The deviation of the battery SOC difference in two adjacent detection periods of the same battery is less than SOC_Th1, otherwise the battery voltage data detection is abnormal; the slope deviation of the battery temperature accumulated value in two adjacent detection periods of the same battery is less than ST_K, otherwise the battery voltage data detection is abnormal; the slope deviation of the battery voltage accumulated value in two adjacent detection periods of the same battery is less than SV_K, otherwise the battery voltage data detection is abnormal.

Images