Battery system

Abstract

A battery system includes a battery module, a thermistor, and an ECU. The thermistor detects the temperature of a first region in the battery module. The ECU calculates an estimated temperature of a second region in the battery module, which has a higher temperature than the first region, by adding a temperature correction value to the thermistor's detection value. The ECU is configured to set the temperature correction value based on the time derivative of the detection value. Furthermore, the ECU sets the temperature correction value so that the rate of increase of the estimated temperature of the second region does not exceed a predetermined ratio greater than zero.

Description

Technical Field

[0001] The present disclosure relates to battery systems. Background Art

[0002] Japanese Patent Application Laid-Open No. 2010-108750 discloses a battery pack input / output control device. The control device includes a battery pack, a temperature sensor, a voltage sensor, a current sensor, and a maximum temperature estimating unit. The battery pack is composed of a plurality of battery cells. The temperature sensor detects the battery surface temperature. The voltage sensor detects the battery voltage of the battery pack's battery cells. The current sensor detects the current values input and output to the battery pack. The maximum temperature estimating unit uses the values detected by the temperature, voltage, and current sensors to estimate the maximum temperature inside the battery pack. Summary of the Invention

[0003] Problems to be solved by the invention

[0004] A battery module comprising a plurality of bipolar nickel-metal hydride batteries is known. In such a battery module, a temperature sensor for detecting its temperature is provided. Here, due to structural reasons of a battery module composed of bipolar batteries, it is sometimes difficult to provide a temperature sensor in a region of the battery module that is the object of temperature analysis (hereinafter referred to as the "target region"). In this case, the temperature sensor is provided in a region different from the target region, and therefore the detection value of the temperature sensor is often inconsistent with the temperature of the target region. Therefore, it is considered to use the detection value of the temperature sensor to estimate the temperature of the target region. In particular, when the temperature of the target region is higher than the region where the temperature sensor is provided, it is important to appropriately estimate the temperature of the target region from the perspective of overheat protection of the battery module.

[0005] Here, because the temperature sensor is affected by noise, its detected value may deviate significantly from the actual temperature of the area where the temperature sensor is installed, increasing. Consequently, the estimated temperature of the target area calculated using the detected value also deviates significantly from the actual temperature of the area, increasing. As a result, the temperature of the target area cannot be accurately estimated.

[0006] In the present disclosure, in a battery system including bipolar nickel-metal hydride batteries, the temperature of a target region having a higher temperature than that of a region where a temperature sensor is installed is appropriately estimated based on a detected value of the temperature of the region.

[0007] Technical solutions to problems

[0008] The battery system disclosed herein includes a battery module, a temperature sensor, and a processing device. The battery module includes a plurality of bipolar nickel-metal hydride batteries. The temperature sensor detects the temperature of a first region in the battery module. The processing device is configured to calculate an estimated temperature of a second region in the battery module, which has a higher temperature than the first region, by adding a temperature correction value to the value detected by the temperature sensor. The processing device is configured to set the temperature correction value based on the time derivative of the detected value. Furthermore, the processing device is configured to set the temperature correction value so that the rate of increase of the estimated temperature of the second region does not exceed a predetermined ratio greater than zero.

[0009] With this configuration, even if the temperature sensor's detection value is affected by noise, causing the time differential value to rise sharply, the rate of increase in the estimated temperature of the second region can be prevented from exceeding a predetermined rate. Consequently, a sudden rise in the estimated temperature of the second region, which would cause the estimated temperature to deviate significantly from the actual temperature, can be suppressed.

[0010] The processing device may be configured to calculate the first estimated temperature of the second region by adding a positive first correction amount to the detected value, and to calculate the second estimated temperature of the second region by performing a smoothing process on the first estimated temperature to make changes in the first estimated temperature more gradual. The processing device may be configured to set the first correction amount to the temperature correction amount when the time differential value is positive. The processing device may be configured to set a second correction amount greater than 0 to the temperature correction amount when the time differential value decreases from a positive value to a negative value, and to calculate the second estimated temperature as the estimated temperature of the second region by adding the second correction amount to the detected value.

[0011] With the above configuration, as the time-differential value of the detected value decreases from a positive value to a negative value, the estimated temperature of the second region changes from the first estimated temperature to a second estimated temperature that is greater than the detected value by a second correction amount. This prevents the estimated temperature of the second region from deviating significantly from the actual temperature due to a sudden decrease in the estimated temperature of the second region to the detected value.

[0012] When the detected value increases from a first value to a second value greater than the first value, the processing device may be configured to calculate a third correction amount obtained by multiplying a time differential value by a predetermined constant; and when the amount of increase from the fourth correction amount indicating the temperature correction amount when the detected value is the first value to the third correction amount is less than a threshold amount, the third correction amount is set to the temperature correction amount when the detected value is the second value. When the amount of increase is greater than the threshold amount, the processing device may be configured to set a value obtained by adding a value less than the threshold amount to the fourth correction amount as the temperature correction amount when the detected value is the second value.

[0013] This configuration prevents the temperature correction amount from increasing excessively even when the temperature sensor is affected by noise, causing the increase to exceed the threshold. Consequently, it prevents the estimated temperature of the second region from increasing excessively due to noise, thereby significantly deviating from the actual temperature.

[0014] The processing device can also be configured to obtain a sampling value of the output of the temperature sensor for each sampling period, use multiple sampling values within a specified period longer than the sampling period, calculate the detection value for each specified period, calculate the estimated temperature of the second area based on the detection value for each specified period, and when the absolute value of the difference between the previous value and the current value of the sampling value is greater than a threshold, the detection value is calculated without using the current value.

[0015] With this configuration, even if the current sampling value is abnormal, such that the absolute value of the difference between the previous sampling value and the current sampling value exceeds the threshold, the processing device does not use the abnormal current sampling value to calculate the detection value. Thus, since the temperature correction amount is set based on the detection value, the estimated temperature of the second region can be appropriately calculated. As a result, it is possible to avoid a situation where the estimated temperature of the second region deviates significantly from the actual temperature.

[0016] The processing device may be configured to set the temperature correction amount so that the rate of increase of the estimated temperature of the second region does not exceed a predetermined rate when the charging current of the battery module is equal to or higher than a predetermined threshold current.

[0017] With this configuration, even when the temperature sensor's detection value is affected by noise due to excessive charging current, the temperature correction value is set so that the rate of increase in the estimated temperature of the second region does not exceed a predetermined rate. As a result, during overcharging of the battery module, a sudden increase in the estimated temperature of the second region, which could significantly deviate from the actual temperature, can be suppressed.

[0018] The battery system may further include a relay unit connected to the battery module. The processing device may be configured to control the relay unit to an OFF state when the estimated temperature of the second region rises to a threshold temperature.

[0019] With this configuration, charging and discharging in the battery module are stopped when the estimated temperature of the second region rises to a threshold temperature. This prevents the temperature of the second region from rising above the threshold temperature. Consequently, the battery module can be protected from overheating caused by charging and discharging.

[0020] In a battery system including bipolar nickel-metal hydride batteries, the temperature of a target region having a higher temperature than that of a region where a temperature sensor is installed can be appropriately estimated based on a detected value of the temperature of the region. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The features, advantages, and technical and industrial significance of exemplary embodiments of the present invention will be described below with reference to the accompanying drawings, wherein like symbols represent like elements, and wherein:

[0022] Figure 1 It is a diagram schematically showing the configuration of a battery system according to this embodiment.

[0023] Figure 2 It is a cross-sectional view of a battery cell.

[0024] Figure 3 This is an exploded perspective view of the battery unit.

[0025] Figure 4 This is a diagram for explaining a situation in which the temperature of a target area near a thermistor is higher than the temperature of an area where the thermistor is provided during use of a battery cell.

[0026] Figure 5 This is a graph showing the temporal changes in the detection value of the thermistor and the estimated temperature of the target area during overcharge of the battery module.

[0027] Figure 6 This is a flowchart showing an example of the steps of the process involved in calculating the estimated temperature of the target area.

[0028] Figure 7 This is a flowchart for explaining the details of a process for setting a temperature correction amount according to a provisional estimated temperature.

[0029] Figure 8 This is a graph showing an example of temporal changes in the detection value of the thermistor and the first estimated temperature of the target area.

[0030] Figure 9 This is a graph showing an example of temporal changes in the detection value of the thermistor and the estimated temperature of the target area in the second embodiment.

[0031] Figure 10 This is a flowchart showing an example of processing executed by the ECU according to the second embodiment.

[0032] Figure 11 This is a graph showing an example of the detection value of the thermistor, the temporary estimated temperature of the target area, and the temporal transition of the estimated temperature of the target area.

[0033] Figure 12This is a graph showing an example of the detection value of the thermistor, the provisional estimated temperature of the target area, and the temporal transition of the estimated temperature of the target area in the third embodiment.

[0034] Figure 13 This is a flowchart showing an example of processing executed by the ECU according to the third embodiment.

[0035] Figure 14 This is a graph showing the temporal transition of the detection value of the thermistor and the estimated temperature of the target area when the sampled value from the thermistor includes an abnormal value.

[0036] Figure 15 This is a graph showing the temporal transition of the detection value of the thermistor and the estimated temperature of the target area when the sampling value of the thermistor includes an abnormal value in the fourth embodiment.

[0037] Figure 16 It is used to illustrate that in the fifth embodiment Figure 6 Flowchart showing the details of the processing executed in step S105. DETAILED DESCRIPTION

[0038] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In the figures, identical or corresponding parts are denoted by the same reference numerals, and their descriptions will not be repeated. The battery system of the present disclosure is typically applied to vehicles, but can also be applied to other applications than vehicles.

[0039] Implementation Method 1

[0040] Figure 1 The structure of the battery system according to the first embodiment is schematically shown. Figure 1 The battery system 90 includes a battery cell 1 , a thermistor 80 , a system main relay (SMR) 95 , a current sensor 53 , and an ECU (Electronic Control Unit) 100 .

[0041] The battery unit 1 is formed by stacking a plurality of battery modules 4. Each battery module 4 is configured to include a bipolar nickel-hydrogen battery. Figure 1 In the example, seven battery modules 4 are described, but the number of battery modules is not limited to this. Cooling plates (not shown) are provided between adjacent battery modules 4, through which a refrigerant for cooling the battery modules 4 flows. The detailed structure of the battery modules 4 will be described later.

[0042] The thermistor 80 detects the temperature of the end region of the battery module 4. In this example, the thermistor 80 is installed in a representative battery module 4 (the middle module among the seven battery modules 4) among the multiple battery modules 4 and detects the temperature of the end region of this battery module. However, the thermistor 80 may also detect the temperature of other battery modules. The detection value of the thermistor 80 is obtained by the ECU 100. Other types of temperature sensors may also be used in place of the thermistor 80.

[0043] The SMR 95 is connected to the battery module 4 (battery cell 1). The SMR 95 is provided as an on / off device for switching the start and stop of charge and discharge in the battery module 4. The current sensor 53 detects the charge and discharge current of the battery module 4 (battery cell 1).

[0044] ECU 100 includes a CPU (Central Processing Unit) 102 and a memory 105. CPU 102 performs various operations based on information stored in memory 105. Memory 105 includes ROM (Read Only Memory) and RAM (Random Access Memory) (neither of which is shown). ROM stores programs executed by CPU 102. RAM temporarily stores data referenced by CPU 102. While control of ECU 100 is implemented through software processing, it can also be implemented through hardware built into ECU 100.

[0045] The ECU 100 estimates the temperature of a target area in the battery module 4 based on various sensor signals (e.g., detection values from the thermistor 80 and the current sensor 53) and programs, data, and maps stored in the memory 105. In the present embodiment, the temperature is estimated in the battery module 4 provided with the thermistor 80. The detected and estimated temperatures in a representative battery module 4 are used as temperatures equivalent to the detected and estimated temperatures in other battery modules 4.

[0046] The ECU 100 acquires sampled values of the thermistor 80 output for each sampling cycle. Using multiple sampled values within a predetermined period longer than the sampling cycle, the ECU 100 calculates the detection value of the thermistor 80 for each predetermined period. Specifically, the ECU 100 calculates the average of the multiple sampled values within the predetermined period as the detection value of the thermistor 80. Based on this detection value, the ECU 100 calculates the estimated temperature of the target area for each predetermined period.

[0047] Furthermore, when the estimated temperature of the target region of thermistor 80 exceeds the threshold temperature, ECU 100 controls SMR 95 to the OFF state. The control by ECU 100 will be described in detail later.

[0048] The battery system 90 is connected to a load 55. When the battery system 90 is mounted on a vehicle, the load 55 is, for example, a PCU (Power Control Unit) and an electric motor.

[0049] Reference Figure 2 and Figure 3 , the structures of the battery unit 1 and the battery module 4 are described. Figure 2 It is a cross-sectional view of the battery cell 1 . Figure 3 It is an exploded perspective view of the battery unit 1.

[0050] The battery unit 1 includes a stacked body 11 and a restraining member 12. The stacked body 11 includes a plurality of battery modules 4. The plurality of battery modules 4 are stacked in a stacking direction Z. The stacked body 11 has end faces 13 and 14 aligned in the stacking direction Z.

[0051] The battery module 4 includes a plurality of electrode layers 41. The electrode layer 41 includes a separator 44, a negative electrode 45, and a positive electrode 43. The separator 44 has an upper surface 46 and a lower surface 47 aligned in the stacking direction Z. The negative electrode 45 is provided on the upper surface 46. The positive electrode 43 is provided on the lower surface 47.

[0052] The battery module 4 also includes a positive terminal electrode and a negative terminal electrode (neither of which is shown in the figure). Therefore, in the battery module 4, a plurality of electrode layers 41 are provided between the positive terminal electrode and the negative terminal electrode.

[0053] A single battery cell is formed between adjacent electrode layers 41. Here, if n electrode layers (bipolar electrodes) 41 are provided in the battery module 4, a total of n-1 battery cells (n is a natural number) are formed between the n electrode layers 41. Furthermore, a single battery cell is formed between the electrode layer 41 adjacent to the positive terminal electrode and the positive terminal electrode. A single battery cell is formed between the electrode layer 41 adjacent to the negative terminal electrode and the negative terminal electrode. Therefore, the battery module 4 includes n+1 battery cells formed in the stacking direction Z. Each battery cell is a bipolar nickel-metal hydride battery.

[0054] The restraining member 12 restrains the stacked body 11 in the stacking direction Z. The restraining member 12 includes a pressing plate 20 , a pressing plate 21 , and a plurality of connecting shafts 26 . The pressing plate 20 presses the end surface 13 . The pressing plate 21 presses the end surface 14 .

[0055] Each connecting shaft 26 is disposed between the pressing plate 20 and the pressing plate 21 to connect the pressing plate 20 and the pressing plate 21. Each connecting shaft 26 is disposed at intervals so as to surround the periphery of the stacked body 11 ( Figure 3 The stacked body 11 is exposed to the outside air, and the heat of the battery modules 4 in the stacked body 11 is released into the outside air.

[0056] Figure 4 This is a diagram for explaining a case where the temperature of a target area near the thermistor 80 is higher than the temperature of the area where the thermistor 80 is provided during use of the battery cell 1 .

[0057] Due to structural reasons of the battery module 4, the thermistor 80 cannot be provided in the target area 107 of the battery module 4, which is the target of temperature analysis. Therefore, the thermistor 80 is provided in the area 85 closest to the target area 107 outside the heat generation area 502 (rectangular area) of the battery module 4. Figure 4 In the example, the region 85 is an end region of the battery module 4 when viewed from the stacking direction Z. During use of the battery module 4 , the temperature of the target region 107 (heat generating region 502 ) is higher than the temperature of the region 85 .

[0058] Due to heat conduction from the target region 107 to the region 85 and the heat capacity of the thermistor 80 , a temperature increase in the target region 107 is not immediately reflected in the detection value of the thermistor 80 .

[0059] For such reasons, the detection value of the thermistor 80 often does not coincide with the temperature of the target region 107. Therefore, it is conceivable to estimate the temperature of the target region 107 using the detection value of the thermistor 80.

[0060] In this embodiment, the ECU 100 calculates the estimated temperature of the target area 107 by adding a temperature correction value to the detection value of the thermistor 80. Specifically, the ECU 100 calculates the estimated temperature TN(t) of the target area 107 at time t based on the detection value TD(t) of the thermistor 80 and the temperature correction value CA(t) using the following equation (1).

[0061] TN(t)=TD(t)+CA(t)…(1)

[0062] The ECU 100 is configured to set the temperature correction amount CA(t) based on the time differential value of the detection value TD(t) of the thermistor 80 .

[0063] like Figure 4 As in the example of , when the temperature of the target area 107 is higher than that of the area 85 , it is important to appropriately estimate the temperature of the target area 107 from the perspective of overheat protection of the battery module 4 .

[0064] Here, the thermistor 80 is sometimes affected by noise (for example, high-frequency noise such as BCI (Bulk Current Injection) noise generated by an external wireless device). As a result, the detection value of the thermistor 80 sometimes deviates significantly from the actual temperature of the area 85 and increases. In more detail, the current flowing through the thermistor 80 is affected by the magnetic field generated by the noise. As a result, the thermistor 80 generates heat, and the detection value of the thermistor 80 deviates significantly from the actual temperature of the area 85 and increases. As a result, the estimated temperature of the target area 107 calculated based on the detection value of the thermistor 80 also deviates significantly from the actual temperature of the target area 107 and increases. As a result, the temperature of the target area 107 cannot be properly estimated.

[0065] The battery system 90 according to this embodiment has a structure for resolving the aforementioned issues. Specifically, the ECU 100 sets the temperature correction amount CA(t) so that the rate of increase over time of the estimated temperature TN(t) of the target region 107 does not exceed a predetermined rate (rate protection value) greater than zero. This process is also referred to as "rate protection processing."

[0066] By executing the ratio guard process, it is possible to suppress the estimated temperature TN(t) from significantly deviating from the actual temperature of the target area 107 and rapidly rising due to the influence of noise. The calculation method of the estimated temperature TN(t) will be described in detail below.

[0067] The provisional estimated temperature of target area 107 at time t is TNP(t). Provisional estimated temperature TNP(t) is used to calculate estimated temperature TN(t) in equation (1). In this first embodiment, estimated temperature TN(t) is calculated by performing the aforementioned ratio protection processing on provisional estimated temperature TNP(t). Provisional estimated temperature TNP(t) is the sum of a provisional temperature correction amount (described later) and detected value TD(t).

[0068] Furthermore, let the actual temperature of region 85 where thermistor 80 is installed at time t be TT1(t). Let the actual temperature of target region 107 at time t be TT2(t). Let the amount of heat transferred between target region 107 and region 85 be W, and let the thermal conductivity be C. Let the heat capacity of thermistor 80 be α. Let the calculation interval (constant value) based on the detection value of thermistor 80 by ECU 100 be Δt. In this case, the following equations (2) and (3) hold true.

[0069] W=C×(TT2(t)-TT1(t))…(2)

[0070] TD(t)=TD(t-Δt)+(W / α)×Δt...(3)

[0071] Here, heat capacity α and thermal conductivity C are constants determined in advance through experiments. The second term on the right side of equation (3) represents the temperature rise of thermistor 80 during Δt due to the heat conduction amount W.

[0072] Here, if TT1(t)=TD(t), TT2(t)=TNP(t), and (α / C) is a constant K, then the following equation (4) holds based on the above equations (1) and (2).

[0073] TNP(t)=TD(t)+((TD(t)-TD(t-Δt)) / Δt)×K

[0074] =TD(t)+TDD(t)×K

[0075] =TD(t)+CAP(t)(CAP(t)≥0)…(4)

[0076] In the above formula (4), TDD(t) is the time differential value of TD(t). CAP(t) is a temporary temperature correction value for the detection value TD(t) of the thermistor 80. The temporary temperature correction value CAP(t) is used to calculate the temperature correction value CA(t) in formula (1). The temporary temperature correction value CAP(t) is set to be equal to or greater than 0. In addition, when the differential value TDD(t) decreases from a positive value to 0, the temporary temperature correction value CAP(t) is set to be equal to the value immediately before the differential value TDD(t) decreases to 0. In the case where the differential value TDD(t) decreases from a positive value to a negative value, the temporary temperature correction value CAP(t) is set to 0.

[0077] The ECU 100 is configured to set the temperature correction amount CA(t) using equations (1) and (4). For example, if the rate of increase of the provisional estimated temperature TNP(t) does not exceed a predetermined rate, the ECU 100 does not execute the rate protection process. Therefore, the ECU 100 sets the provisional temperature correction amount CAP(t) to the temperature correction amount CA(t) (CA(t) = CAP(t)). As a result, the provisional estimated temperature TNP(t) is used as the estimated temperature TN(t) (TN(t) = TNP(t)).

[0078] On the other hand, if the rate of increase of the provisional estimated temperature TNP(t) exceeds a predetermined ratio, the ECU 100 executes a rate protection process. In this case, the ECU 100 does not set the provisional temperature correction amount CAP(t) itself as the temperature correction amount CA(t). As a result, the provisional estimated temperature TNP(t) itself is not used as the estimated temperature TN(t). By executing the rate protection process in this manner, it is possible to prevent the estimated temperature TN(t) from significantly deviating from the actual temperature of the target area 107 due to noise and causing a sudden increase. In this embodiment, if the rate of increase of the provisional estimated temperature TNP(t) exceeds a predetermined ratio, the ECU 100 executes the rate protection process so that the rate of increase of the estimated temperature TN(t) remains within the predetermined ratio.

[0079] Figure 5 Graph showing the temporal changes of the detection value TD(t) of the thermistor 80 and the estimated temperature TN(t) of the target area 107 during overcharge of the battery module 4. Figure 5 , the horizontal axis represents time and the vertical axis represents temperature.

[0080] In addition, "when the battery module 4 is overcharged" means when the charging current of the battery module 4 is equal to or higher than a predetermined threshold current. The threshold current is appropriately determined in advance through experiments, etc., in order to protect the battery module 4 from overheating. When the battery module 4 is overcharged, the heating area 502 ( Figure 4 ) is particularly prone to rapid temperature rise in the target area 107. In such a case where the temperature of the target area 107 is particularly prone to rapid temperature rise, it is particularly effective for the ECU 100 to calculate the estimated temperature TN(t) using equations (1) and (4) (performing ratio protection processing). In addition, the ECU 100 can calculate the estimated temperature TN(t) based on the current sensor 53 ( Figure 1 ) is used to determine whether the battery module 4 is overcharged.

[0081] exist Figure 5 In FIG. 6 , line 605 shows the temporal change of the detection value TD(t) of the thermistor 80. Line 610 shows the temporal change of the estimated temperature TN(t) of the target area 107 when the thermistor 80 is not affected by noise. Line 615 shows the temporal change of the estimated temperature TN(t) of the target area 107 when the thermistor 80 is affected by noise.

[0082] At time t10, the battery module 4 starts to charge, and the temperature of the battery module 4 starts to rise. After time t10, the detection value TD(t) (line 605) of the thermistor 80 does not match the estimated temperature TN(t) (line 610) of the target area 107. Specifically, Figure 5As shown, during overcharging of the battery module 4, the estimated temperature TN(t) rises faster than the detection value TD(t) of the thermistor 80. It is experimentally known that the actual temperature of the target area 107 rises faster than the detection value TD(t). Therefore, the estimated temperature TN(t) of the target area 107 appropriately reflects the actual temperature of the target area 107.

[0083] Assume that at time t11, the detection value TD(t) of the thermistor 80 begins to be affected by noise. As a result, the rate of increase of the provisional estimated temperature TNP(t), calculated from the detection value TD(t) of the thermistor 80 using equation (4), exceeds a predetermined rate PDR (rate protection value). The ECU 100 sets the temperature correction amount CA(t) (line 615) so that the rate of increase of the estimated temperature TN(t) reaches the predetermined rate PDR. That is, after time t11, the estimated temperature TN(t) is represented by line 615. The predetermined rate PDR is set based on the rate of increase that would be achieved during overcharging if the thermistor 80 were not affected by noise (the rate of increase of line 610 after time t11). For example, the predetermined rate PDR is set to the maximum value of this rate of increase. This prevents the estimated temperature TN(t) from being affected by noise and rising to a level significantly deviating from the actual temperature of the target area 107 during overcharging of the battery module 4.

[0084] At time t12, the estimated temperature TN(t) of the target area 107 reaches the threshold temperature THT. In response, the ECU 100 switches the SMR 95 from the closed state to the open state to protect the battery module 4 from overheating. The threshold temperature THT is a value used to protect the battery module 4 from overheating and is appropriately determined in advance through experiments, etc. Switching the SMR 95 to the open state stops charging the battery module 4. This prevents the estimated temperature TN(t) of the target area 107 from rising above the threshold temperature THT. As a result, the battery module 4 can be protected from overheating caused by overcharging.

[0085] Figure 6 This is a flowchart showing an example of processing steps associated with calculating the estimated temperature TN(t) of the target area 107. This flowchart is executed at predetermined time intervals during startup of a driving system of a vehicle equipped with the battery system 90 (while the vehicle's starter switch is on).

[0086] Reference Figure 6ECU 100 calculates the detection value TD(t) of thermistor 80 (step S105). Specifically, ECU 100 uses multiple sampling values within a predetermined period longer than the sampling cycle to calculate the detection value corresponding to the predetermined period. ECU 100 calculates the differential value TDD(t) of the detection value TD(t) (step S110).

[0087] The ECU 100 calculates the temporary temperature correction amount CAP(t) (=TDD(t)×K) in equation (4) based on the differential value TDD(t) of the detected value TD(t) (step S115). Using equation (4), the ECU 100 calculates the temporary estimated temperature TNP(t) of the target area 107 based on the detected value TD(t) and the temporary temperature correction amount CAP(t) (step S120).

[0088] The ECU 100 sets the temperature correction amount CA(t) based on the provisional estimated temperature TNP(t) (step S125). The details of this setting method will be described later.

[0089] The ECU 100 adds the set temperature correction amount CA(t) to the detection value TD(t) according to the equation (1) to thereby calculate the estimated temperature TN(t) of the target area 107 (step S130 ).

[0090] In step S135, the ECU 100 determines whether the estimated temperature TN(t) is equal to or higher than the threshold temperature THT. If the estimated temperature TN(t) is lower than the threshold temperature THT ("No" in step S135), the ECU 100 controls the SMR 95 to a closed state (step S140). The process then returns to the return state while the battery cell 1 continues to charge and discharge.

[0091] On the other hand, if the estimated temperature TN(t) is equal to or higher than the threshold temperature THT ("YES" in step S135), the ECU 100 controls the SMR 95 to the OFF state (step S145). The ECU 100 controls the SMR 95 to the OFF state until the estimated temperature TN(t) decreases below the threshold temperature THT (until the control branch in step S135 returns to "NO").

[0092] Figure 7 This is a process for explaining the process of setting the temperature correction amount CA(t) based on the temporary estimated temperature TNP(t). Figure 6 Flowchart of the detailed contents of step S125).

[0093] Reference Figure 7ECU 100 then determines whether the rate of increase of the provisional estimated temperature TNP(t) is greater than or equal to the prescribed rate PDR (step S205). If the rate of increase of the provisional estimated temperature TNP(t) is greater than or equal to the prescribed rate PDR ("YES" in step S205), ECU 100 sets the temperature correction amount CA(t) so that the rate of increase of the estimated temperature TN(t) in target area 107 does not exceed the prescribed rate PDR (step S210). In this case, the provisional temperature correction amount CAP(t) is not used as the temperature correction amount CA(t). Specifically, ECU 100 sets the temperature correction amount CA(t) so that the rate of increase of the estimated temperature TN(t) reaches the prescribed rate PDR.

[0094] On the other hand, if the rate of increase of the provisional estimated temperature TNP(t) is less than the predetermined rate PDR ("No" in step S205), the ECU 100 sets the provisional temperature correction amount CAP(t) to the temperature correction amount CA(t) (step S215). After steps S210 and S215, the ECU 100 ends. Figure 7 Processing, so that the processing enters Figure 6 Step S130.

[0095] As described above, the ECU 100 according to the present embodiment calculates the estimated temperature TN(t) of the target area 107 by adding the temperature correction value CA(t) to the detection value TD(t) of the thermistor 80. The ECU 100 is configured to set the temperature correction value CA(t) based on the time differential value TDD(t) of the detection value TD(t) of the thermistor 80. The ECU 100 sets the temperature correction value CA(t) so that the rate of increase of the estimated temperature TN(t) of the target area 107 over time does not exceed a predetermined ratio PDR greater than zero.

[0096] With this configuration, even if the detection value TD(t) of thermistor 80 is affected by noise and the differential value TDD(t) rises sharply, the rate of increase of the estimated temperature TN(t) of the target area 107 can be prevented from exceeding the predetermined rate PDR. Consequently, a sudden rise that would cause the estimated temperature TN(t) of the target area 107 to deviate significantly from the actual temperature can be suppressed. Consequently, the temperature of the target area 107 can be appropriately estimated based on the detection value TD(t) of thermistor 80.

[0097] Then, the timing at which the SMR 95 is controlled to be in the OFF state as the estimated temperature TN(t) reaches the threshold temperature THT is reached (at the time of Figure 5 That is, it is possible to prevent the battery module 4 from being excessively protected or insufficiently protected from overheating, thereby enabling the battery module 4 to be appropriately protected.

[0098] Furthermore, in this embodiment, the calculation of the estimated temperature TN(t) by the ECU 100 does not require the detection values of the current sensor and the voltage sensor. In other words, the ECU 100 can calculate the estimated temperature TN(t) of the target area 107 based on the detection value TD(t) of the thermistor 80, without using the current and voltage input and output to the battery module 4.

[0099] Implementation Method 2

[0100] In the second embodiment, the case where the differential value TDD(t) of the detection value TD(t) decreases from a positive value to a negative value is described. Specifically, the ECU 100 performs a smoothing process on the estimated temperature of the target area 107 based on the calculation method before the differential value TDD(t) decreases to a negative value. Then, the ECU 100 calculates the temperature after the smoothing process as the estimated temperature TN(t) of the target area 107 after the differential value TDD(t) decreases to a negative value. The smoothing process is a gradual process including averaging and delaying. The structure and processing steps of the battery system in the second embodiment are the same as those in the second embodiment. Figures 1 to 6 The structure and processing steps of the battery system 90 in the illustrated embodiment 1 are basically the same.

[0101] Reference Figure 8 , a comparative example relative to this embodiment 2 is described. Figure 8 This is a graph showing an example of the time-dependent change of the detection value TD(t) of the thermistor 80 and the first estimated temperature of the target area 107. The first estimated temperature is a temperature calculated from the detection value TD(t) using the above equations (1) and (4) without smoothing, under the condition that the rate of increase of the estimated temperature TN(t) does not exceed the prescribed rate PDR (rate protection value). Figure 8 , the horizontal axis represents time and the vertical axis represents temperature.

[0102] A line 210 shows the temporal transition of the detection value TD(t) of the thermistor 80. A line 205 shows the temporal transition of the first estimated temperature of the target region 107.

[0103] At time t21, thermistor 80's detection value TD(t) begins to be affected by noise. The ECU sets the temperature correction value CA(t) (line 205) so that the rate of increase of the estimated temperature TN(t) does not exceed the predetermined rate PDR. This temperature increase continues for a period ΔT1. During this period, thermistor 80 is affected by noise. Consequently, thermistor 80's detection value TD(t) is higher than the actual temperature in region 85.

[0104] At time t22, the rise in detection value TD(t) stops. Then, the differential value TDD(t) of detection value TD(t) decreases from a positive value to a negative value. As a result, as described below, estimated temperature TN(t) rapidly decreases to the detection value TD(t) of thermistor 80 (line 205).

[0105] The temperature correction amount CA(t) is set based on the temporary temperature correction amount CAP(t) and is therefore dependent on the differential value TDD(t) of the detection value TD(t) (Equations (1) and (4)). In the above-mentioned equation (4), it is assumed that when the differential value TDD(t) decreases to a negative value, the temporary temperature correction amount CAP(t) is set to 0. Therefore, during the period ΔT2 during which the detection value TD(t) decreases, the temperature correction amount CA(t) is also set to 0 (CA(t) = CAP(t) = 0). Therefore, the estimated temperature TN(t) of the target area 107 is equal to the detection value TD(t) of the thermistor 80 (Equation (1)). For this reason, at time t22, the estimated temperature TN(t) of the target area 107 drops sharply to the detection value TD(t) of the thermistor 80.

[0106] However, since the actual temperature of the target area 107 may not be as Figure 8 Therefore, the estimated temperature TN(t) of the target area 107 does not appropriately reflect the actual temperature of the target area 107. Thus, from the perspective of the accuracy of the estimated temperature TN(t), the phenomenon of the estimated temperature TN(t) dropping (changing) sharply is not preferable.

[0107] At time t23, the influence of noise on thermistor 80 disappears, and the detection value TD(t) of thermistor 80 begins to rise again (line 210). As the detection value TD(t) rises, the differential value TDD(t) of the detection value TD(t) becomes positive. As a result, the temperature correction value CA(t) calculated based on the differential value TDD(t) also becomes positive.

[0108] During the period from time t23 to time t24, the rate of increase of the provisional estimated temperature TNP(t) is set to be greater than or equal to the prescribed rate PDR. Therefore, the ECU sets the temperature correction amount CA(t) so that the rate of increase of the estimated temperature TN(t) does not exceed the prescribed rate PDR (limiting the rate of increase of the estimated temperature TN(t) to the prescribed rate PDR). As a result, it takes time for the estimated temperature TN(t) of the target area 107 to appropriately reflect the actual temperature (catch up with the actual temperature).

[0109] As mentioned above, Figure 8The comparative example shown has the following problems: the accuracy of the estimated temperature TN(t) of the object area 107 is reduced (time t22 to time t24); and after the detection value TD(t) is no longer affected by noise, it takes time for the estimated temperature TN(t) to properly reflect the actual temperature again (time t23 to time t24).

[0110] Figure 9 This is a graph showing an example of temporal changes in the detection value TD(t) of the thermistor 80 and the estimated temperature TN(t) of the target area 107 in the second embodiment.

[0111] Reference Figure 9 Line 255 shows the temporal change of the first estimated temperature of target area 107. Line 260 shows the temporal change of the second estimated temperature of target area 107. The second estimated temperature is calculated by ECU 100 performing a smoothing process on the first estimated temperature to make the change of the first estimated temperature gentle.

[0112] The time constant of the smoothing process is based on the approximate generation period of the noise (for example, Figure 8 The time constant is determined by the sum of the lengths of the period ΔT1 and the period ΔT2. As an example, the time constant is a number smaller than the sum and greater than 0. By determining the time constant in this manner, the second estimated temperature of the target area 107 appropriately reflects the actual temperature of the target area 107.

[0113] At time t21, Figure 8 Similarly, the detection value TD(t) of thermistor 80 begins to be affected by noise. The first estimated temperature, representing the estimated temperature TN(t) of the target area 107, begins to rise so that its rate of increase does not exceed the predetermined rate PDR (line 255). In this second embodiment, when the differential value TDD(t) is positive (during the period until time t22), the ECU 100 sets a positive correction value CA1(t) to the temperature correction value CA(t). The ECU 100 calculates the first estimated temperature, which is the estimated temperature TN(t) of the target area 107, by adding the correction value CA1(t) to the detection value TD(t) of thermistor 80. Here, the ECU 100 calculates the second estimated temperature together with the first estimated temperature (line 260). That is, during the period until time t22, the first estimated temperature is used as the estimated temperature TN(t) of the target area 107, but the second estimated temperature is also calculated.

[0114] At time t22, the rise in the detection value TD(t) of thermistor 80 stops (line 210). Then, the differential value TDD(t) of the detection value TD(t) decreases from a positive value to a negative value. Consequently, the estimated temperature TN(t) of target area 107 decreases rapidly.

[0115] Here, after time t22, as the estimated temperature TN(t) of the target area 107, the second estimated temperature (line 260) is used instead of the first estimated temperature. Specifically, the ECU 100 sets a correction amount CA2(t) greater than 0 for the temperature correction amount CA(t). The ECU 100 calculates the second estimated temperature as the estimated temperature TN(t) of the target area 107 by adding the correction amount CA2(t) to the detection value TD(t) of the thermistor 80. In this way, after time t22, the second estimated temperature is used as the estimated temperature TN(t), so that at time t22, the estimated temperature TN(t) of the target area 107 drops to the second estimated temperature. Therefore, the estimated temperature TN(t) drops sharply at time t22, which is different from the estimated temperature ( Figure 8 205 ), but unlike the case of the comparative example, it does not decrease to the detection value TD(t) of the thermistor 80 (line 210 ).

[0116] After time t22, the rate of increase of the estimated temperature TN(t) (second estimated temperature) of the target area 107 increases without being restricted by the predetermined rate PDR. When the estimated temperature TN(t) of the target area 107 does not increase sharply in this way, it is considered that it appropriately reflects the actual temperature of the target area 107. Therefore, in this embodiment 2, compared with the case of the comparative example ( Figure 8 Compared with the time t22 to t24 of FIG. 2 , the accuracy of the estimated temperature TN(t) is improved. Figure 8 Unlike time t23 to time t24 (in FIG. 1 ), it takes no time for the estimated temperature TN(t) of the target area 107 to appropriately reflect the actual temperature after it drops sharply.

[0117] Figure 10 This is a flowchart showing an example of processing executed by the ECU 100 according to the second embodiment. Figure 7 Compared with the flowchart of Figure 7 ) and execute the processing of steps S315, S320, and S325. Figure 10 The processing of steps S305 and S310 in the flowchart of Figure 7 The processing of steps S205 and S210 is the same.

[0118] Reference Figure 10If the rate of increase of the provisional estimated temperature TNP(t) is less than the prescribed rate PDR ("No" in step S305), the ECU 100 determines whether the differential value TDD(t) of the detection value TD(t) of the thermistor 80 has decreased from a positive value to a negative value (step S315). If the differential value TDD(t) has not decreased from a positive value to a negative value ("No" in step S315), the ECU 100 sets the correction value CA1(t) to the temperature correction value CA(t) (step S320). As described above, the correction value CA1(t) is Figure 9 On the other hand, when the differential value TDD(t) decreases from a positive value to a negative value ("Yes" in step S315), the ECU 100 sets the correction amount CA2(t) to the temperature correction amount CA(t) (step S325). As described above, the correction amount CA2(t) is set before the time t22. Figure 9 After steps S320 and S325, ECU 100 ends Figure 10 Processing, so that the processing enters Figure 6 Step S130.

[0119] As described above, the ECU 100 according to the second embodiment calculates the first estimated temperature of the target area 107 by adding the positive correction value CA1(t) to the detection value TD(t). The ECU 100 then calculates the second estimated temperature of the target area 107 by smoothing the first estimated temperature to make its change more gradual. When the differential value TDD(t) is positive, the ECU 100 sets the correction value CA1(t) to the temperature correction value CA(t). When the differential value TDD(t) decreases from a positive value to a negative value, the ECU 100 sets the correction value CA2(t) greater than zero to the temperature correction value CA(t). The ECU 100 then adds the correction value CA2(t) to the detection value TD(t) to calculate the second estimated temperature, which is the estimated temperature TN(t) of the target area 107.

[0120] With the above configuration, as the differential value TDD(t) decreases from a positive value to a negative value, the estimated temperature TN(t) of the target area 107 changes from the first estimated temperature to the second estimated temperature that is greater than the detected value TD(t) by the correction amount CA2(t). This prevents the estimated temperature TN(t) of the target area 107 from deviating significantly from the actual temperature due to a sudden decrease in the estimated temperature TN(t) of the target area 107 to the detected value TD(t).

[0121] Implementation 3

[0122] In this third embodiment, a case is described where the increase in the temporary temperature correction amount CAP(t) (=TDD(t)×K)) is greater than a threshold value greater than 0. In this case, it is not preferable that the temperature correction amount CA(t) is set too high and the estimated temperature TN(t) is calculated as an unrealistic value that is too high. Therefore, in this third embodiment, the temperature correction amount CA(t) is set so as to avoid such a situation. This point will be described in detail below. The structure and processing steps of the battery system in this third embodiment are similar to those in the embodiment 3. Figures 1 to 6 The structure and processing steps of the battery system 90 in the illustrated embodiment 1 are basically the same.

[0123] First, refer to Figure 11 , a comparative example relative to this embodiment 3 is described. Figure 11 1 is a graph showing an example of the temporal transition of the detection value TD(t) of the thermistor 80, the provisional estimated temperature TNP(t) of the target region 107, and the estimated temperature TN(t) of the target region 107. Figure 11 , the horizontal axis represents time and the vertical axis represents temperature.

[0124] Line 315 shows the temporal change of the detection value TD(t) of thermistor 80. Line 305 shows the temporal change of the provisional estimated temperature TNP(t) calculated using equation (4). Line 310 shows the temporal change of the estimated temperature TN(t) of target area 107. Line 320 shows the temporal change of the provisional temperature correction amount CAP(t) calculated using equation (4). Line 325 shows the temporal change of the temperature correction amount CA(t). After time t33 (described later), lines 320 and 325 coincide.

[0125] At time t31 , the detection value TD(t) of the thermistor 80 starts to rise from the temperature T0 due to the influence of noise (line 315 ).

[0126] At time t32, the provisional estimated temperature TNP(t) rises sharply due to the influence of noise (line 305). Figure 11 In the example shown in FIG3 , the temporal rate of increase of the provisional estimated temperature TNP(t) exceeds a predetermined rate PDR (rate protection value). Therefore, between time t32 and time t33, the ECU sets the temperature correction amount CA(t) (lines 310 and 325) so that the rate of increase of the estimated temperature TN(t) does not exceed the predetermined rate PDR.

[0127] At time t33, the increase in temperature correction value CA(t) and estimated temperature TN(t) completes. Estimated temperature TN(t) reaches temperature T1. After time t33, the detected value TD(t) (temperature TA) of thermistor 80 increases by ΔTE1 compared to temperature T0 before thermistor 80 was affected by noise (before time t31). Furthermore, the temperature correction value CA(t) increases by an increase amount CAQ.

[0128] exist Figure 11 In the example, it is assumed that the detection value TD(t) of thermistor 80 increases by ΔTE1 due to noise from time t31 to time t33 during charging of battery module 4. That is, it is assumed that the actual temperature of target area 107 does not increase by ΔTE1.

[0129] As the detected value TD(t) rises sharply by ΔTE1, the temporary temperature correction amount CAP(t) calculated based on the detected value TD(t) also rises sharply (line 320). The increase amount CAQ of the temporary temperature correction amount CAP(t) is greater than the threshold amount THQ. The threshold amount THQ is a value that is unrealistic for the increase amount CAQ to exceed the threshold amount THQ, and is appropriately determined in advance through experiments, etc. Therefore, if Figure 11 As shown, when the increase amount CAQ of the temporary temperature correction amount CAP(t) is equal to or greater than the threshold amount THQ, it is considered that the actual temperature in the target area 107 has not increased sharply and the thermistor 80 is affected by noise.

[0130] However, in Figure 11 In the comparative example, the temporary temperature correction amount CAP(t) is assumed to have increased by an unrealistic increase amount CAQ, the temperature correction amount CA(t) is calculated, and the estimated temperature TN(t) is calculated. As a result, the estimated temperature TN(t) excessively increases to an unrealistic temperature T1 (line 310). Therefore, in Figure 11 In the comparative example of , the estimated temperature TN(t) of the target region 107 does not appropriately reflect the actual temperature of the target region 107. Therefore, the ECU cannot appropriately calculate the estimated temperature TN(t).

[0131] Therefore, in this third embodiment, the ECU 100 determines whether the thermistor 80 is affected by noise based on whether the increase CAQ of the temporary temperature correction value CAP(t) is greater than the threshold value THQ. If the increase CAQ is greater than the threshold value THQ, the ECU 100 limits the increase in the temporary temperature correction value CAP(t) so that the temporary temperature correction value CAP(t) increases by an amount less than the threshold value THQ. The temperature correction value CA(t) is then set based on the limited increase. This prevents the temperature correction value CA(t) from being set excessively high. Consequently, it prevents the estimated temperature TN(t) from rising to an unrealistic temperature T1. This will be explained in detail below.

[0132] Reference Figure 12 Next, the processing by ECU 100 according to the third embodiment will be described. Figure 12 1 is a graph showing an example of the temporal transition of the detection value TD(t) of the thermistor 80, the provisional estimated temperature TNP(t) of the target region 107, and the estimated temperature TN(t) of the target region 107 in the third embodiment. Figure 12 , the horizontal axis represents time and the vertical axis represents temperature.

[0133] Line 415 shows the temporal change of the detection value TD(t) of thermistor 80. Line 405 shows the temporal change of the provisional estimated temperature TNP(t) calculated using equation (4). Line 410 shows the temporal change of the estimated temperature TN(t) of target area 107. Line 420 shows the temporal change of a provisional temperature correction amount (described later) whose increase relative to provisional temperature correction amount CAP(t) is limited. Line 425 shows the temporal change of temperature correction amount CA(t) according to Embodiment 3. After time t43 (described later), lines 420 and 425 coincide.

[0134] At time t41 , the detection value TD(t) of the thermistor 80 starts to rise from the temperature T0 due to the influence of noise (line 415 ).

[0135] At time t42, the increase CAQ of the temporary temperature correction amount CAP(t) becomes greater than the threshold amount THQ due to the influence of noise. Therefore, the ECU 100 determines that the thermistor 80 is affected by noise. Here, the ECU 100 limits the increase of the temporary temperature correction amount CAP(t) to CAQ1 (<CAQ) (line 420). The increase CAQ1 is less than the threshold amount THQ and greater than 0. The increase CAQ1 is obtained, for example, by dividing the increase CAQ by a specified number greater than 1. The temporary temperature correction amount after the increase is limited is set to CAPR(t). In this embodiment, the temporary temperature correction amount CAPR(t) is used in place of the temporary temperature correction amount CAP(t) in equations (1) and (4).

[0136] Furthermore, at time t42, the temporal rate of increase of the provisional estimated temperature TNP(t) calculated based on the provisional temperature correction value CAPR(t) exceeds a predetermined rate PDR (rate protection value) (line 405). Therefore, between time t42 and time t43, the ECU 100 increases and sets the temperature correction value CA(t) so that the rate of increase of the estimated temperature TN(t) does not exceed the predetermined rate PDR (lines 410 and 425).

[0137] At time t43, the increase in the temperature correction amount CA(t) since time t42 reaches the increase amount CAQ1. After time t43, the ECU 100 sets the correction amount CA3 (constant value) to the temperature correction amount CA(t) and does not increase the temperature correction amount CA(t). As a result, the estimated temperature TN(t) becomes lower than that of the comparative example ( Figure 11 ) is smaller than the temperature T1 in the case of the temperature T2 (line 410). Here, the detection value TD(t) of the thermistor 80 is different from that of the comparative example ( Figure 11 ), the temperature of thermistor 80 has increased by ΔTE1 compared to before thermistor 80 was affected by noise (before time t41). Unlike the comparative example, however, the temperature correction amount CA(t) is set by ECU 100 so that it is increased by only the increase amount CAQ1 compared to before time t41. Specifically, the correction amount CA3 set after time t42, when the increase amount CAQ of the provisional temperature correction amount CAP(t) exceeds the threshold amount THQ, is the sum of the correction amount CA4 set before time t42 and the increase amount CAQ1 (line 425).

[0138] Figure 13 This is a flowchart showing an example of processing executed by the ECU 100 according to the third embodiment. Figure 6 The detailed processing of step S125 is as follows. Figure 13 Flowchart and Figure 7Compared with the flowchart of FIG, the processing of steps S402, S403, and S404 is added. Figure 13 The processing of steps S405, S410, and S415 in the flowchart of Figure 7 The processing of steps S205, S210, and S215 is the same.

[0139] and Figure 13 Refer to it together Figure 12 The ECU 100 determines whether the increase CAQ of the temporary temperature correction amount CAP(t) is greater than or equal to the threshold amount THQ (step S402). If the increase CAQ of the temporary temperature correction amount CAP(t) is greater than or equal to the threshold amount THQ ("YES" in step S402), the ECU 100 calculates the limited increase CAQ1 (step S403) and calculates the temporary temperature correction amount CAPR(t) with limited increase based on the increase CAQ1 and the pre-increase correction amount CA4 (step S404). The temporary temperature correction amount CAPR(t) with limited increase is used in place of the temporary temperature correction amount CAP(t) in subsequent processing. After step S404, the ECU 100 proceeds to step S405. On the other hand, if the increase CAQ of the temporary temperature correction amount CAP(t) is less than the threshold amount THQ ("NO" in step S402), the ECU 100 does not limit the increase of the temporary temperature correction amount CAP(t) and proceeds to step S405.

[0140] The processing after step S405 is the same as that in the case of embodiment 1 ( Figure 7 ) is executed in the same manner. Specifically, when the rate of increase of the temporary estimated temperature TNP(t) calculated based on the temporary temperature correction amount CAP(t) (or the temporary temperature correction amount CAPR(t)) is greater than the prescribed ratio PDR ("Yes" in step S405), the ECU 100 sets the temperature correction amount CA(t) in such a way that the rate of increase of the temporary estimated temperature TNP(t) of the target area 107 does not exceed the prescribed ratio PDR (step S410). If this is not the case ("No" in step S405), the ECU 100 sets the temporary temperature correction amount CAP(t) (or the temporary temperature correction amount CAPR(t)) to the temperature correction amount CA(t) (step S415). After steps S410 and S415, the ECU 100 ends Figure 13 Processing, so that the processing enters Figure 6 Step S130.

[0141] As described above, in this embodiment, when the detected value TD(t) rises from temperature T0 to temperature TA, the ECU 100 calculates the temporary temperature correction amount CAP(t) by multiplying the differential value TDD(t) by a predetermined constant K. If the increase CAQ from the correction amount CA4, which represents the temperature correction amount CA(t) when the detected value TD(t) is at temperature T0, to the temporary temperature correction amount CAP(t) is less than the threshold amount THQ, the ECU 100 sets the temporary temperature correction amount CAP(t) to the temperature correction amount CA(t) when the detected value TD(t) is at temperature TA. On the other hand, if the increase CAQ is greater than the threshold amount THQ, the ECU 100 sets the temperature correction amount CA(t) when the detected value TD(t) is at temperature TA by adding the increase CAQ1, which is less than the threshold amount THQ, to the correction amount CA4 as the temperature correction amount CA(t) when the detected value TD(t) is at temperature TA.

[0142] Thus, when the thermistor 80 is affected by noise and the temporary temperature correction amount CAP(t) rises by an amount CAQ exceeding the threshold value THQ, it is possible to avoid the temperature correction amount CA(t) being set too high. Figure 11 Unlike the case of the comparative example (line 310 ), it is possible to avoid the estimated temperature TN(t) of the target region 107 from excessively rising to an unrealistic temperature T1 (line 410 ).

[0143] Implementation 4

[0144] In the fourth embodiment, when the absolute value of the difference between the previous value and the current value of the sampled value obtained from the thermistor 80 is greater than the threshold value, the ECU 100 calculates the detection value TD(t) without using the current value. The structure and processing steps of the battery system in the fourth embodiment are similar to those in the embodiment 1. Figures 1 to 6 The structure and processing steps of the battery system 90 in the illustrated embodiment 1 are basically the same.

[0145] First, refer to Figure 14 , a comparative example relative to this embodiment 4 is described. Figure 14 Graph 1 is a graph showing the temporal transition of the detection value TD(t) of the thermistor 80 and the estimated temperature TN(t) of the target area 107 when the sampled value from the thermistor 80 includes an abnormal value. Figure 14 In the figure, the horizontal axis represents time and the vertical axis represents temperature.

[0146] Line 510 shows the temporal change in the detection value TD(t) of thermistor 80. Line 505 shows the temporal change in the estimated temperature TN(t) of target area 107. Line 505T shows the temporal change in the temperature considered to be the actual temperature of target area 107. As described above, Δt is the calculation interval based on the detection value of thermistor 80 by ECU 100.

[0147] ECU 100 obtains sampled values from thermistor 80 at a predetermined sampling period. This sampling period is shorter than the detection period of length Δt. ECU 100 calculates the average value of a plurality of (e.g., a predetermined number of) sampled values within the detection period of length Δt as the detection value TD(t) for the next detection period.

[0148] For example, the ECU 100 calculates the average value of a plurality of sample values acquired during the detection period between time t51 and time t52, and uses this average value as the detection value TD(t) during the detection period between time t52 and time t53 to calculate the estimated temperature TN(t) (line 505).

[0149] exist Figure 14 In the example, it is assumed that at least one of the multiple sampled values obtained during the detection period between time t51 and time t52 is an abnormal value. An "abnormal value" is a value that has changed by more than a specified threshold value from the sampled value immediately before the value was sampled (the previous value). Such an abnormal value is not caused by an actual temperature change in region 85 of thermistor 80, but rather by, for example, a sudden misdetection by thermistor 80. Thus, because the sampled values during the period between time t51 and time t52 include an abnormal value, at time t52, the detection value TD(t), which is the average value of the multiple sampled values during that period, drops sharply to VA2 (<VA1).

[0150] Therefore, at time t52, the differential value TDD(t) of the detection value TD(t) (=(VA2-VA1) / Δt) decreases to a negative value. As described above, when the differential value TDD(t) decreases to a negative value, the temperature correction value CA(t) is set to 0. Therefore, during the period from time t52 to time t53, the temperature correction value CA(t) is set to 0. As a result, the estimated temperature TN(t) drops rapidly to a level that matches the detection value TD(t) of thermistor 80 (line 505).

[0151] The phenomenon of the estimated temperature TN(t) changing rapidly in this manner is considered unrealistic. Therefore, it is considered that the estimated temperature TN(t) does not appropriately reflect the actual temperature of the target area 107 during the period from time t52 to time t53.

[0152] During the period from time t53 to time t54, the ECU calculates the estimated temperature TN(t) of the target area 107 (line 505) by using the average of multiple sampled values acquired during the previous period (the period from time t52 to time t53) as the detected value TD(t) for the current period. These multiple sampled values are assumed to exclude any abnormal values. Furthermore, during the period from time t53 to time t54, the rate of increase of the provisional estimated temperature TNP(t) is assumed to exceed the prescribed rate PDR. Therefore, the ECU sets the temperature correction amount CA(t) to limit the rate of increase of the estimated temperature TN(t) to the prescribed rate PDR (rate protection value). As a result, it takes time for the estimated temperature TN(t) of the target area 107 to appropriately reflect the actual temperature (line 505T) (catch up with the actual temperature).

[0153] As mentioned above, in Figure 14 In the comparative example, since the multiple sampling values obtained during the period from time t51 to time t52 include abnormal values, the estimated temperature TN(t) of the target area 107 during the period from time t52 to time t55 cannot appropriately reflect the temperature considered to be the actual temperature of the target area 107 (line 505T).

[0154] Reference Figure 15 A method for calculating the detection value TD(t) by the ECU 100 according to the fourth embodiment will be described. Figure 15 1 is a graph showing the temporal changes of the detection value TD(t) of the thermistor 80 and the estimated temperature TN(t) of the target area 107 when the sampled value of the thermistor 80 includes an abnormal value in the fourth embodiment. Figure 15 In the figure, the horizontal axis represents time and the vertical axis represents temperature.

[0155] A line 605 shows the temporal transition of the estimated temperature TN(t) in the fourth embodiment. A line 610 shows the temporal transition of the detection value TD(t) in the fourth embodiment.

[0156] If the absolute value of the difference between the previous and current sampled values is greater than a threshold, ECU 100 calculates detection value TD(t) without using the current value. The threshold is used by ECU 100 to determine whether the current value is abnormal based on the difference between the previous and current values. Specifically, the threshold is appropriately predetermined to a value that is unrealistic for the amount of change in the sampled value per sampling cycle.

[0157] During the period from time t51 to time t52, it is assumed that at the time point when ECU 100 obtains the current value, the difference between the previous value and the current value is greater than the threshold value (the current value is an abnormal value). Here, ECU 100 replaces the current value, which is an abnormal value, with the previous value, which is a normal value. Then, ECU 100 calculates the average value of the multiple sampling values after replacing the abnormal value with the normal value as the detection value TD(t) (line 610). Figure 15 In the example, the average value is VA3 (≠VA2). Based on the average value VA3 of the multiple sampling values during the period from time t51 to time t52, the ECU 100 calculates the detection value TD(t) corresponding to the period from time t52 to time t53. Then, the ECU 100 sets the temperature correction amount CA(t) from time t52 to time t53 based on the detection value TD(t). As a result, Figure 15 In the example of , the estimated temperature TN(t) during this period is calculated to be equal to the estimated temperature TN(t) at time t52.

[0158] Thus, in this embodiment 4, compared with the case of the comparative example ( Figure 14 Unlike line 505 (shown in Figure 5), this prevents abnormal sampling values from being reflected in the calculation results of estimated temperature TN(t) during the next period (between time t52 and time t53). As a result, it prevents the estimated temperature TN(t) from changing rapidly at time t52 (line 605).

[0159] In addition, Figure 15 In the example above, the case where the sampled value drops sharply is described. However, if the sampled value rises sharply (when the abnormal value is significantly larger than the other sampled values), the ECU 100 calculates the detection value TD(t) in the same manner as described above. Specifically, the ECU 100 calculates the detection value TD(t) based on the other sampled values, without using the abnormal value. The ECU 100 then sets the temperature correction amount CA(t) based on the detection value TD(t) to calculate the estimated temperature TN(t) of the target area 107.

[0160] Figure 16 This is used to explain the Figure 6 Flowchart showing the details of the processing executed in step S105.

[0161] Reference Figure 16ECU 100 obtains a sampled value from thermistor 80 (step S505). ECU 100 determines whether the absolute value of the difference between the current sampled value and the previous sampled value is greater than threshold value TH (step S510). If the absolute value is not greater than threshold value TH ("No" in step S510), ECU 100 proceeds to step S520. If the absolute value is greater than threshold value TH ("Yes" in step S510), ECU 100 deems the current sampled value to be abnormal and replaces the current sampled value with the previous sampled value (step S515).

[0162] In step S520, the ECU 100 determines whether a predetermined detection period (in Figure 15 In the example, the period of length Δt). If the predetermined detection period has not passed ("No" in step S520), ECU100 returns the process to step S505. If the predetermined detection period has passed ("Yes" in step S520), ECU100 calculates the average value of the plurality of sampling values after the abnormal value is replaced with the normal value (step S525). This average value is used as the detection value of the thermistor 80. Figure 6 It is used in the processing after step S110.

[0163] As described above, in the fourth embodiment, the ECU 100 determines whether the absolute value of the difference between the previous value and the current value of the sampled value obtained from the thermistor 80 is greater than or equal to the threshold value TH. If the absolute value is greater than or equal to the threshold value TH, the ECU 100 determines that the current value is an abnormal value and calculates the detection value TD(t) using the previous value (and other normal values during the predetermined detection period) instead of the current value.

[0164] Thus, even if the current value is abnormal, such that the absolute value of the difference between the previous sampled value and the current value exceeds the threshold value TH, the current value, which is an abnormal value, is not used to calculate the detected value TD(t). This prevents the abnormal value of the sampled value from being reflected in the temperature correction amount CA(t). As a result, it is possible to prevent the estimated temperature TN(t) of the target area 107 from deviating significantly from the actual temperature due to a sudden change.

[0165] Modification 1

[0166] In each of the above-described embodiments, the thermistor 80 is provided only in a representative battery module 4 among the plurality of battery modules 4. Alternatively, a thermistor may be provided in each of the plurality of battery modules 4. The detection value of each thermistor is acquired by the ECU 100.

[0167] In this case, the ECU 100 calculates the estimated temperature TN(t) of the target area 107 for each battery module 4. Furthermore, the ECU 100 may also calculate the estimated temperature TN(t) of the target area 107 based on whether the average value of the seven estimated values (or at least one of these estimated values) of the estimated temperature TN(t) calculated for each battery module 4 is the threshold temperature THT( Figure 5 ) above, to make step S135 ( Figure 6 ) processing branch.

[0168] Modification 2

[0169] In each of the above-described embodiments, the detection value TD(t) of the thermistor 80 is used to calculate the estimated temperature TN(t) of the target region 107 (nearby region) of the thermistor 80 .

[0170] In contrast, the detection value TD(t) of the thermistor 80 can also be used to estimate the temperature of an area other than the target area 107 of the thermistor 80 (another area in the heat generation area 502). In other words, as long as the thermal conductivity (heat transfer amount) between the area where the thermistor 80 is installed and the other area whose temperature is to be estimated is known, the ECU 100 can calculate an estimated value of the temperature of the other area using equations (1) and (4).

[0171] Other variations

[0172] In the above-described first to fourth embodiments, as shown in equations (1) and (4), the ECU 100 calculates the differential value TDD(t) every time the calculation interval Δt (constant value) of the detection value TD(t) of the thermistor 80 passes.

[0173] On the other hand, the ECU 100 may execute the process of calculating the differential value TDD(t) every time the detection value TD(t) of the thermistor 80 changes by a predetermined temperature interval (referred to as ΔTI).

[0174] In this case, the smaller ΔTI is, the more likely the differential value TDD(t) is to fluctuate. On the other hand, the larger ΔTI is, the less likely the differential value TDD(t) is to fluctuate, but the number of calculations decreases, or the time until the next calculation of the differential value TDD(t) increases. Alternatively, the ECU 100 can increase the number of calculations of the differential value TDD(t) using the method described below.

[0175] Specifically, the ECU 100 divides the calculation process of the differential value TDD(t) into n steps (n is a natural number greater than or equal to 2) while staggering its start time. For example, the ECU 100 first starts the first differential value calculation process. This process is completed at the time when the detection value TD(t) of the thermistor 80 increases by ΔTI after the start of this process. The differential value in this process is calculated based on the time-dependent rate of change between the detection value TD(t) of the thermistor 80 at the start of this process and the value at which the detection value TD(t) increases by ΔTI.

[0176] Each time the detection value TD(t) of thermistor 80 increases by ΔTI×(m-1) / n from the detection value TD(t) at the start of the first differential value calculation process, ECU 100 starts the mth differential value calculation process (m is a natural number satisfying 1≤m≤n). Each differential value calculation process is completed when the detection value TD(t) of thermistor 80 increases by ΔTI from the start of the process.

[0177] Thus, after the first differential value calculation process is completed, the corresponding differential value calculation process is completed at each timing at which the detected value TD(t) of thermistor 80 rises by a temperature of ΔTI / n. Therefore, the ECU 100 can obtain the differential value TDD(t) of the detected value TD(t) of thermistor 80 at each of these timings. As a result, the ECU 100 can prevent significant fluctuations in the differential value TDD(t) and increase the number of times the differential value TDD(t) is calculated compared to a case where the differential value TDD(t) calculation process is not performed in a split manner. Consequently, the accuracy of the temperature estimation of the target area 107 can be improved.

[0178] The embodiments disclosed herein are to be considered in all respects as illustrative and non-restrictive. The scope of the present invention is not indicated by the above description but by the scope of the claims, and is intended to include all modifications within the scope and meaning equivalent to the scope of the claims.

Claims

1. A battery system, characterized in that: have: a battery module comprising a plurality of bipolar nickel-metal hydride batteries; a temperature sensor, configured to detect a temperature of a first region in the battery module; and a processing device configured to calculate an estimated temperature of a second region of the battery module having a temperature higher than that of the first region by adding a temperature correction amount to a detection value of the temperature sensor; The processing device is configured to set the temperature correction amount according to a time differential value of the detection value, set the temperature correction amount so that the rate of increase of the estimated temperature of the second region does not exceed a predetermined ratio greater than zero, calculate a first estimated temperature of the second region by adding a positive first correction amount to the detection value, and calculate a second estimated temperature of the second region by performing a smoothing process on the first estimated temperature to make the change of the first estimated temperature gentle. The processing device is configured to determine whether the rate of increase of the provisional estimated temperature of the second region is equal to or greater than the predetermined rate. When the rate of increase of the provisional estimated temperature is greater than or equal to the predetermined rate, the temperature correction amount is set so that the rate of increase of the estimated temperature of the second region does not exceed the predetermined rate. On the other hand, when the rate of increase of the provisional estimated temperature is less than the predetermined rate, it is determined whether the time differential value decreases from a positive value to a negative value. If the time differential value does not decrease from a positive value to a negative value, the temperature correction amount is set to the first correction amount, the first correction amount being a value obtained by multiplying the time differential value by a predetermined positive constant; When the time differential value decreases from a positive value to a negative value, a second correction amount greater than 0 is set for the temperature correction amount, and the second estimated temperature as the estimated temperature of the second region is calculated by adding the second correction amount to the detected value.

2. The battery system according to claim 1, wherein: When the detection value increases from a first value to a second value greater than the first value, the processing device is configured to: calculating a third correction amount obtained by multiplying the time differential value by the predetermined constant; When an increase from a fourth correction amount indicating the temperature correction amount when the detection value is the first value to the third correction amount is less than a threshold value, setting the third correction amount to the temperature correction amount when the detection value is the second value; and When the increase amount is equal to or greater than the threshold amount, a value obtained by adding a value smaller than the threshold amount to the fourth correction amount is set as the temperature correction amount when the detection value is the second value.

3. The battery system according to claim 1 or claim 2, characterized in that: The processing device is composed of: Obtaining a sample value of the output of the temperature sensor for each sampling period, The detection value is calculated for each predetermined period using a plurality of sampling values within a predetermined period longer than the sampling cycle. For each of the predetermined periods, an estimated temperature of the second region is calculated based on the detected value. When the absolute value of the difference between the previous value and the current value of the sampling value is greater than or equal to a threshold value, the current value is not used in calculating the detection value.

4. The battery system according to claim 3, characterized in that The processing device is configured to calculate an average value of the sampling values as the detection value.

5. The battery system according to claim 1 or claim 2, characterized in that: The processing device is configured to set the temperature correction amount so that the rate of increase of the estimated temperature of the second region does not exceed the predetermined ratio when the charging current of the battery module is equal to or greater than a predetermined threshold current.

6. The battery system according to claim 1 or claim 2, characterized in that: The battery system further includes a relay unit connected to the battery module. Here, the processing device is configured to control the relay unit to an OFF state when the estimated temperature of the second region rises to a threshold temperature.

7. The battery system according to claim 6, characterized in that The relay unit is configured to switch between execution and stop of charge and discharge in the battery module.

Images