A battery temperature early warning method for an on-board terminal of an intelligent network-connected vehicle

Abstract

The present invention relates to the field of automobile battery monitoring technology, and specifically to a battery temperature early warning method for an on-board terminal of an intelligent networked automobile. The method includes: obtaining a temperature sequence and a voltage sequence to be analyzed of a battery when the automobile is driving; obtaining the state bit of each voltage data of the voltage sequence, and forming a state bit sequence, segmenting the state bit sequence, and thereby achieving the purpose of segmenting the temperature sequence to be analyzed; establishing a window with a temperature data as the center to calculate the interference coefficient of the temperature data; obtaining the intensity coefficient of each temperature data showing a sharp upward trend according to the difference and slope value between each two adjacent data in the temperature sequence to be analyzed, and then obtaining the prediction weight of the temperature data; based on the prediction weight of each temperature data, using the ARIMA algorithm to obtain the predicted temperature at the time to be predicted, and performing a vehicle battery temperature early warning. The present application can improve the accuracy of vehicle battery temperature prediction and perform real-time battery temperature early warning.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile battery monitoring, and in particular to a method for early warning of battery temperature at an on-board terminal of an intelligent network-connected automobile. Background Art

[0002] The power batteries of new energy vehicles will generate heat when working. Under normal circumstances, the heat is controllable, but when the battery temperature is too high or the charging voltage is too high, chemical reactions inside the power battery will occur one after another, resulting in a chain reaction, causing the internal pressure and temperature of the battery to rise sharply, and then causing thermal runaway of the battery, leading to combustion or explosion.

[0003] Traditional prediction of power battery temperature generally collects battery temperature time series data through temperature sensors, uses ARIMA algorithm (Autoregressive Integrated Moving Average Model) to calculate predicted temperature, and issues an alarm when battery temperature reaches or exceeds the set threshold. The weight distribution in the ARIMA algorithm is mostly equal weight distribution, but this method loses the historical nature of the data and cannot reflect the sharp rise in temperature. At the same time, the power battery may be disturbed by external factors during monitoring, resulting in abnormal sensor readings. The weight of such disturbed data should be appropriately reduced. The equal weight distribution of the traditional ARIMA algorithm will affect the accuracy of vehicle battery temperature prediction, and thus affect the operation of the vehicle's battery temperature control system. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle. The technical solution adopted is as follows:

[0005] An embodiment of the present invention provides a method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle, the method comprising:

[0006] Obtain the temperature data sequence and voltage data sequence of the battery when the car is running; obtain the lag order using the temperature data sequence, and obtain the temperature sequence to be analyzed according to the lag order and the temperature data sequence;

[0007] The voltage data at the same time as each data of the temperature sequence to be analyzed is obtained in the voltage data sequence to form a voltage sequence; the state bit of each voltage data is obtained by using the difference between every two adjacent data in the voltage sequence;

[0008] The state bits of each voltage data are combined into a state bit sequence, and the state bit sequence is segmented to obtain different state bit data segments; the state bit data segments are mapped to the temperature sequence to be analyzed to obtain temperature data segments;

[0009] A window is established with a temperature data as the center in a temperature data segment, and the interference coefficient of the temperature data is calculated according to the variance of the difference between every two adjacent data in the window and the data before and after the temperature data;

[0010] According to the difference and slope value between each two adjacent data in the temperature sequence to be analyzed, the intensity coefficient of each temperature data showing a sharp upward trend is obtained; according to the intensity coefficient and interference coefficient of the temperature data showing a sharp upward trend, the prediction weight of the temperature data is obtained;

[0011] The predicted temperature at the time to be predicted is obtained using the ARIMA algorithm based on the prediction weight of each temperature data in the temperature sequence to be analyzed; the vehicle battery temperature warning is performed based on the predicted temperature.

[0012] Preferably, obtaining a temperature data sequence and a voltage data sequence of a battery when the vehicle is running includes:

[0013] The battery management system BMS is used to collect real-time temperature data and voltage data of the vehicle driving battery to form a temperature data sequence and a voltage data sequence respectively, and the collection time interval is a preset interval.

[0014] Preferably, the hysteresis order is obtained by using the temperature data sequence, and the temperature sequence to be analyzed is obtained according to the hysteresis order and the temperature data sequence, including:

[0015] The ACF and PACF diagrams are drawn using the collected temperature data sequence, from which the lag order P is obtained; according to the lag order, the last P temperature data in the temperature data sequence are selected to form the temperature sequence to be analyzed.

[0016] Preferably, obtaining the state bit of each voltage data by using the difference between every two adjacent data in the voltage sequence includes:

[0017] The status bit is represented as 0 or 1, 0 represents the discharge state, and 1 represents the charge state. The difference between every two adjacent voltage data in the voltage sequence is calculated, that is, the latter voltage data minus the previous voltage data. If the difference is greater than or equal to zero, the status bit of the latter voltage data in the adjacent voltage data is 1. If the difference is less than zero, the status bit of the latter voltage data in the adjacent voltage data is 0.

[0018] Preferably, segmenting the state bit sequence to obtain different state bit data segments includes:

[0019] Traverse from the first state bit of the state bit sequence. If a data segment with the same state bits and the number of the same state bits is greater than or equal to the preset number is found during the traversal process, it is recorded as the starting segment. The traversal starts again from the next state bit of the starting segment. If a data segment with a state bit different from the starting segment is found, but the state bits are also the same and the number of the same state bits is greater than or equal to the preset number, it is recorded as the ending segment. The traversal stops, and the state bits between the starting segment and the previous data of the ending segment constitute the state bit data segment. Similarly, the traversal continues from the ending segment. At this time, the ending segment is the starting segment of the next state bit data segment, and the state bit data segments in the state bit sequence are obtained.

[0020] Preferably, mapping the state bit data segment to the temperature sequence to be analyzed to obtain the temperature data segment includes:

[0021] The position of each state bit data segment in the state bit sequence is mapped to the temperature sequence to be analyzed, the temperature sequence to be analyzed is segmented, and the temperature data segment corresponding to each state bit data segment is obtained.

[0022] Preferably, the calculation formula of the interference coefficient of temperature data is:

[0023] ,

[0024] in, Represents the interference coefficient of the i-th temperature data in the temperature sequence to be analyzed; represents the variance of the difference between every two adjacent data in the window centered on the i-th temperature data in the temperature sequence to be analyzed; m represents the number of temperature data in the window; represents the variance of the difference between every two adjacent data in the m+ temperature data after the i-th temperature data in the temperature data segment to which the i-th temperature data belongs; represents the variance of the difference between every two adjacent data in the m-temperature data before the i-th temperature data in the temperature data segment to which the i-th temperature data belongs; It represents the distance between the i-th temperature data and a straight line fitted by using other data except the i-th temperature data in the temperature data segment to which the i-th temperature data belongs.

[0025] Preferably, the intensity coefficient of each temperature data showing a sharp upward trend is obtained according to the difference and slope value between each two adjacent data in the temperature sequence to be analyzed, including:

[0026] The difference between every two adjacent data in the temperature sequence to be analyzed is used to form a difference sequence, the difference is the subtraction of the previous data from the next data, and the difference between every two adjacent data is recorded as the difference corresponding to the next data; if the difference corresponding to a temperature data in the temperature sequence to be analyzed is less than or equal to zero, then the intensity coefficient of the temperature data showing a sharp upward trend is zero; if the difference corresponding to a temperature data in the temperature sequence to be analyzed is greater than zero, then the slope between the temperature data and the previous temperature data is taken as the slope term, the absolute value of the difference between the difference corresponding to the temperature data and the maximum value in the difference sequence is obtained, and the difference difference term is obtained by inverting it, and the average of the slope term and the difference difference term is the intensity coefficient of the temperature data showing a sharp upward trend.

[0027] Preferably, the prediction weight of the temperature data is obtained according to the intensity coefficient and the interference coefficient of the temperature data showing a sharp upward trend, including:

[0028] Calculate the difference between the preset value and the interference coefficient of a temperature data and multiply it by the intensity coefficient of the sharp upward trend of the temperature data to obtain the credibility of the sharp upward trend of the temperature data; calculate the inverse of the time interval between the moment corresponding to the temperature data and the moment to be predicted, which is recorded as the degree of proximity; calculate the average value of the credibility of the sharp upward trend of the temperature data and the degree of proximity of the temperature data and normalize them to obtain the prediction weight of the temperature data.

[0029] Preferably, the vehicle battery temperature warning is performed according to the predicted temperature, including:

[0030] Set the battery temperature threshold. If the predicted temperature is greater than or equal to the battery temperature threshold, an early warning is issued.

[0031] The embodiment of the present invention has at least the following beneficial effects: the present invention obtains a temperature sequence and a voltage sequence to be analyzed, and obtains the state bit of each voltage data by using the difference between each two adjacent data in the voltage sequence, so as to achieve the purpose of state division of the vehicle battery, and then divides the temperature sequence to be analyzed according to the divided state to obtain temperature data segments, so that the subsequent analysis of the temperature data is all in one state, and the temperature data is analyzed in combination with the characteristics of the charge and discharge state, thereby increasing the accuracy of the analysis; further, the interference coefficient of the temperature data and the intensity coefficient of each temperature data showing a sharp upward trend are calculated in one temperature data segment, and the prediction weight of each temperature data in the temperature sequence to be analyzed is determined by combining these two characteristics, and the predicted temperature at the time to be predicted is obtained by using the ARIMA algorithm based on the prediction weight of each temperature data in the temperature sequence to be analyzed, and finally the vehicle battery temperature warning is performed, and when performing temperature prediction, the history of the temperature data, the controllability of the change speed and the possibility of interference of the temperature data are combined to give each data in the temperature sequence to be analyzed a corresponding prediction weight, so that the prediction result is more accurate, and a warning can be given before the temperature change is uncontrollable to prevent accidents caused by excessive battery temperature. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0033] Figure 1 A method flow chart of a method for early warning of battery temperature of an on-board terminal of an intelligent connected vehicle provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation method, structure, features and effects of a battery temperature warning method for a vehicle terminal of an intelligent networked vehicle proposed by the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

[0035] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0036] The following is a detailed description of a method for early warning battery temperature of an on-board terminal of an intelligent network-connected vehicle provided by the present invention in conjunction with the accompanying drawings.

[0037] During the driving process of the car, the battery charging and discharging state will change dynamically. At the same time, as the battery is used for a long time, the chemical reaction inside the battery will accelerate and generate more heat. Therefore, the overall battery temperature shows an upward trend, and the data growth rate varies. The data shows many twists and turns. Therefore, it is necessary to monitor the battery temperature, issue an early warning before the battery temperature rises sharply and uncontrollably, and cool the battery.

[0038] Since the sensor is inside the battery pack, when the car drives near certain electrical equipment (such as motors, high-voltage lines, etc.), a short electromagnetic pulse may be generated, and electromagnetic interference affects the accuracy of the sensor monitoring data; the battery box or plug-in may cause insulation failure due to short-term water ingress due to rain, which in turn affects temperature monitoring. When the water evaporates, the insulation failure will be eliminated and the temperature monitoring will return to normal. The data caused by these interferences have obvious fluctuation characteristics, but because the battery temperature itself is a relatively volatile data, it is difficult to distinguish the fluctuation characteristics. However, under normal circumstances, the change in battery temperature is related to the battery's charge and discharge status. However, since the battery temperature data is easily affected, and the battery voltage data changes steadily within a fixed range, the battery's charge and discharge status can be analyzed based on the voltage data at adjacent moments except the target moment, and then combined with the charge and discharge status to analyze the changes in battery temperature in the charging state and the changes in battery temperature in the discharge state.

[0039] Example:

[0040] The main application scenarios of the present invention are: collecting driving battery temperature data of an intelligent networked electric vehicle equipped with a vehicle-mounted system while driving, and improving the problems caused by using average weights when using the ARIMA algorithm for temperature prediction, so as to obtain more accurate temperature prediction data and early warning.

[0041] See also Figure 1 , which shows a method flow chart of a method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle provided by an embodiment of the present invention, the method comprising the following steps:

[0042] Step S1, obtaining a temperature data sequence and a voltage data sequence of a battery when the vehicle is running; obtaining a hysteresis order using the temperature data sequence, and obtaining a temperature sequence to be analyzed according to the hysteresis order and the temperature data sequence.

[0043] The main purpose of this application is to analyze the status and temperature data of the vehicle battery to obtain the prediction weight of each temperature data, and then use the ARIMA algorithm to predict the vehicle battery temperature based on the prediction weight, so as to realize the monitoring and early warning of the vehicle battery temperature.

[0044] Therefore, it is necessary to collect the temperature data of the battery. During the driving process of the electric vehicle, the temperature of the driving battery is mainly related to the charging and discharging of the battery. When the battery is in the charging state, the battery voltage increases and the battery temperature rises rapidly. When the battery is in the discharging state, the battery voltage decreases and the battery temperature rises slowly. The car will enter the charging state in the following situations: when the car is braking or decelerating during driving, the energy recovery system will start, converting the kinetic energy generated by the vehicle during braking or deceleration into electrical energy and storing it in the battery; for some hybrid electric vehicles, when the battery power is low, the engine will start and drive the generator to charge the battery.

[0045] The real-time temperature data and voltage data of the vehicle driving battery are collected by the battery management system BMS which is specially used for monitoring the battery in the vehicle system of the electric vehicle. The collection time interval can be set to a preset interval, preferably once every 2 seconds in the embodiment of the present invention. The collection deadline is the current time.

[0046] The real-time temperature data and voltage data are collected by the temperature sensor and voltage sensor connected to the battery management system BMS, thereby obtaining the real-time temperature data and voltage data from the start of the vehicle to the current moment, which constitute the temperature data sequence and voltage data sequence respectively.

[0047] Furthermore, according to the steps of the ARIMA algorithm, it is necessary to use the data in the collected temperature data sequence to draw ACF and PACF diagrams to obtain the lag order P, and then select the last P temperature data in the temperature data sequence according to the lag order P to form the temperature sequence to be analyzed for subsequent analysis. It should be noted that the arrangement order of the temperature data in the temperature sequence to be analyzed is the time series order.

[0048] It should also be noted that the core component of the drive battery is the battery pack, which contains multiple battery modules, each of which is composed of multiple single cells and a battery management system BMS. In this solution, temperature data and voltage data of single cells are collected for analysis.

[0049] Step S2, obtaining voltage data at the same time as each data of the temperature sequence to be analyzed in the voltage data sequence to form a voltage sequence; and obtaining the state bit of each voltage data by using the difference between every two adjacent data in the voltage sequence.

[0050] In order to determine the charge and discharge status of the battery, and then analyze the temperature data in combination with the charge and discharge status, it is necessary to obtain voltage data with the same timing as the temperature sequence to be analyzed obtained in step S1, and then analyze it. Therefore, the voltage data at the same time as the data of the temperature sequence to be analyzed is obtained in the voltage data sequence to form a voltage sequence.

[0051] Furthermore, it is necessary to judge the charge and discharge state of the battery according to the voltage sequence, and obtain the state bit of each voltage data by using the difference between every two adjacent data in the voltage sequence, the state bit is represented as 0 or 1, 0 represents the discharge state, and 1 represents the charge state, and calculate the difference between every two adjacent voltage data in the voltage sequence, the difference between every two adjacent voltage data is the difference between the latter voltage data and the former voltage data, if the difference is greater than or equal to zero, then the state bit of the latter voltage data in every two adjacent voltage data is 1, if the difference is less than zero, then the state bit of the latter voltage data in every two adjacent voltage data is 0; the state bit of the first voltage data is obtained by using the first voltage data and the voltage data before it. In this way, the state bit of each voltage data in the voltage sequence can be obtained.

[0052] Step S3, the state bits of each voltage data are grouped into a state bit sequence, and the state bit sequence is segmented to obtain different state bit data segments; the state bit data segments are mapped to the temperature sequence to be analyzed to obtain temperature data segments.

[0053] The charging and discharging of the vehicle battery is a process change, so the voltage sequence needs to be segmented according to the state bit of each voltage data in the voltage sequence. Specifically, the state bits of each voltage data are arranged in the order of the voltage data in the voltage sequence to obtain a state bit sequence; traverse from the first state bit of the state bit sequence, if the traversal process finds a data segment with the same state bits and the number of the same state bits is greater than or equal to a preset number, it is recorded as the starting segment, and the traversal starts again from the next state bit of the starting segment. If a data segment with a different state bit from the starting segment is found, but the state bits are also the same and the number of the same state bits is greater than or equal to the preset number, it is recorded as the ending segment, and the traversal stops. The state bits between the starting segment and the previous data of the ending segment constitute a state bit data segment; and so on, continue to traverse from the ending segment, and the ending segment is now the starting segment of the next state bit data segment, and obtain each state bit data segment in the state bit sequence.

[0054] The value of the preset number is 3 in the embodiment of the present invention, and the implementer can adjust it according to the actual situation. In addition, when the state bit sequence is segmented, theoretically, the first few state bits of the state bit sequence may be different from the characteristics of the state bit data segment obtained above. For example, the first few state bits are 01101010. When the starting segment is found after 01101010, the first few state bits 01101010 are a state bit data segment. Theoretically, the last few state bits of the state bit sequence may also have such a situation. At this time, the last few state bits constitute a state bit data segment. However, in actual situations, the charging and discharging process is a process quantity, and the probability of such a situation is extremely small.

[0055] Furthermore, the position of each status bit data segment in the status bit sequence is mapped to the temperature sequence to be analyzed, the segmentation of the temperature sequence to be analyzed is completed, and the temperature data segment corresponding to each status bit data segment is obtained; thereby completing the segmentation of the temperature sequence to be analyzed according to the charging and discharging state of the battery.

[0056] Step S4, establishing a window with a temperature data as the center in a temperature data segment, and calculating the interference coefficient of the temperature data according to the variance of the difference between every two adjacent data in the window and the data before and after the temperature data.

[0057] In step S3, the temperature sequence to be analyzed is segmented according to the charging and discharging state of the vehicle battery. Further, the possibility of interference of each temperature data can be analyzed in each temperature data segment.

[0058] A window is established with a temperature data as the center in a temperature data segment, and the interference coefficient of the temperature data is calculated according to the variance of the difference between every two adjacent data in the window and the data before and after the temperature data. Specifically, the calculation formula of the interference coefficient of the temperature data is as follows:

[0059] ,

[0060] in, Represents the interference coefficient of the i-th temperature data in the temperature sequence to be analyzed; represents the variance of the difference between every two adjacent data in the window centered on the i-th temperature data in the temperature sequence to be analyzed; m represents the number of temperature data in the window; represents the variance of the difference between every two adjacent data in the m+ temperature data after the i-th temperature data in the temperature data segment to which the i-th temperature data belongs; represents the variance of the difference between every two adjacent data in the m-temperature data before the i-th temperature data in the temperature data segment to which the i-th temperature data belongs; It represents the distance between the i-th temperature data and a straight line fitted by using other data except the i-th temperature data in the temperature data segment to which the i-th temperature data belongs.

[0061] In addition, m+ is less than or equal to the number of data after the i-th temperature data in the temperature data segment to which the i-th temperature data belongs, and m- is less than or equal to the number of data before the i-th temperature data in the temperature data segment to which the i-th temperature data belongs.

[0062] Indicates the difference between the data change trend in the window centered on the i-th temperature data and the change trend of the surrounding data. If the difference is small, it means that the change trend of the selected i-th temperature data is consistent with the temperature change trend during the charging or discharging period, and the possibility of interference is small; The smaller the distance, the more likely it is that under ideal conditions (no other factors affecting temperature change), the data changes in the temperature data segment where the ith temperature data is located have certain regularity, and the ith temperature data satisfies this regularity. α represents the interference coefficient of the selected temperature data. If the selected data meets the temperature change characteristics of the period and the more it meets the regularity of temperature data change, the less likely the temperature data is to be interfered.

[0063] Thus, the interference coefficient of each temperature data in the temperature sequence to be analyzed can be obtained.

[0064] Step S5, obtaining the intensity coefficient of each temperature data presenting a sharp upward trend according to the difference and slope value between every two adjacent data in the temperature sequence to be analyzed; obtaining the prediction weight of the temperature data according to the intensity coefficient and interference coefficient of the temperature data presenting a sharp upward trend.

[0065] As the battery is used for a long time, the battery temperature has a clear upward trend. The speed of battery temperature rise is a process from slow to rapid. When the battery temperature rises too quickly, it is difficult to cool down in a short time, which causes the battery temperature to rise to a dangerous temperature value in a short time or even instantly, resulting in battery explosion, etc. Therefore, at each moment, the data with a large temperature change rate should be paid special attention to. If the data point has a strong temperature change rate, it means that the temperature may rise sharply in a short period of time. Therefore, when predicting, it is necessary to pay attention to whether the data shows a sharp upward trend.

[0066] The intensity coefficient of each temperature data showing a sharp upward trend is obtained according to the difference and slope value between every two adjacent data in the temperature series to be analyzed.

[0067] The difference between every two adjacent data in the temperature sequence to be analyzed is used to form a difference sequence, the difference is the subtraction of the previous data from the next data, and the difference between every two adjacent data is recorded as the difference corresponding to the next data; if the difference corresponding to a temperature data in the temperature sequence to be analyzed is less than or equal to zero, then the intensity coefficient of the temperature data showing a sharp upward trend is zero; if the difference corresponding to a temperature data in the temperature sequence to be analyzed is greater than zero, then the slope between the temperature data and the previous temperature data is taken as the slope term, the absolute value of the difference between the difference corresponding to the temperature data and the maximum value in the difference sequence is obtained, and the difference difference term is obtained by inverting it, and the average of the slope term and the difference difference term is the intensity coefficient of the temperature data showing a sharp upward trend.

[0068] When obtaining the difference in the difference sequence and the slope of every two adjacent data, the difference or slope corresponding to the first temperature data in the temperature sequence to be analyzed needs to be obtained using a data before the first temperature data, or interpolated using the calculated differences and slopes corresponding to other data in the sequence to be analyzed.

[0069] The specific calculation formula for the intensity coefficient of the temperature data showing a sharp upward trend is:

[0070] ,

[0071] in, Indicates the intensity coefficient that the i-th temperature data in the temperature sequence to be analyzed presents a sharp upward trend; represents the difference value corresponding to the i-th temperature data in the temperature sequence to be analyzed; K represents the slope value between the i-th temperature data and the previous temperature data; It represents the maximum value of the difference corresponding to all temperature data in the temperature sequence to be analyzed, that is, the maximum value in the difference sequence, and | | represents the absolute value symbol. It represents the difference between the difference value corresponding to the i-th temperature data in the temperature sequence to be analyzed and the maximum difference value in the difference sequence.

[0072] like If it is a positive number, the intensity coefficient analysis of the sharp upward trend is performed, otherwise no analysis is performed, and the intensity coefficient of the sharp upward trend of the i-th temperature data is 0; if the slope value K is larger, it means that the temperature data changes faster and the probability of showing a sharp upward trend is higher; if The smaller it is, the faster the temperature data changes, and the greater the probability of a sharp upward trend. Therefore, the corresponding intensity coefficient of the temperature data showing a sharp upward trend is greater.

[0073] Thus, the intensity coefficient of each temperature data in the temperature series to be analyzed showing a sharp upward trend can be obtained.

[0074] If the vehicle's battery temperature data is less likely to be disturbed, the data will be more valuable for reference. If the temperature data shows a more obvious sharp upward trend, the data will be more worthy of attention. At the same time, based on the historicity of the data, the prediction weight of each temperature data in the temperature data sequence to be analyzed can be comprehensively obtained. A higher prediction weight will be given to temperature data that is less likely to be disturbed, shows an obvious sharp upward trend, and is closer to the temperature at the time to be predicted.

[0075] The prediction weight of the temperature data is obtained according to the intensity coefficient and interference coefficient of the temperature data showing a sharp upward trend. The difference between the preset value and the interference coefficient of a temperature data is calculated and multiplied by the intensity coefficient of the temperature data showing a sharp upward trend to obtain the credibility of the temperature data showing a sharp upward trend; the inverse of the time interval between the moment corresponding to the temperature data and the moment to be predicted is calculated, which is recorded as the degree of proximity; the average value of the credibility of the temperature data showing a sharp upward trend and the degree of proximity of the temperature data is calculated and normalized to obtain the prediction weight of the temperature data.

[0076] The specific calculation formula for the prediction weight of temperature data is:

[0077] ,

[0078] Among them, ω represents the prediction weight of a temperature data, norm represents normalization, and α represents the interference coefficient of the temperature data. The intensity coefficient indicates that the temperature coefficient shows a sharp upward trend. It indicates the credibility of the sharp upward trend of the temperature data, C indicates the time interval between the corresponding time of the temperature data and the time to be predicted, Indicates the degree of proximity. When the temperature data presents a sharp upward trend, the greater the credibility, the greater the degree of proximity, and the greater the prediction weight of the temperature data, that is, the lower the interference coefficient, the greater the intensity coefficient of the temperature data presenting a sharp upward trend, and the closer to the time to be predicted, the greater the prediction weight of the temperature data. In this way, the prediction weight of each temperature data in the temperature sequence to be analyzed can be obtained.

[0079] Step S6, using the ARIMA algorithm based on the prediction weight of each temperature data in the temperature sequence to be analyzed to obtain the predicted temperature at the time to be predicted; and issuing a vehicle battery temperature warning according to the predicted temperature.

[0080] In step S5, the prediction weight of each temperature data in the temperature sequence to be analyzed is obtained, and then the battery temperature data is predicted according to the prediction weight to obtain the battery temperature prediction value at the future moment, and an early warning is issued according to the set threshold.

[0081] Specifically, the predicted temperature at the time to be predicted is obtained by using the ARIMA algorithm in combination with the prediction weight of each temperature data and each temperature data in the temperature sequence to be analyzed; the battery temperature threshold is set, preferably, the battery temperature threshold in the embodiment of the present invention is 55°C, if the predicted temperature is greater than or equal to 55°C, an early warning is issued, the behavior that will cause the battery to enter the charging state is stopped, and measures such as adjusting the charging and discharging power and operating the water cooling system are taken. If necessary, the high-voltage relay can be disconnected, or other measures can be taken to protect the battery safety.

[0082] To sum up, the present application can determine the battery status of a car based on the voltage sequence, analyze each data in the temperature sequence to be analyzed in each battery status and obtain the prediction weight of each temperature data, predict the temperature at future times based on the prediction weight, and issue an alarm when the predicted temperature reaches the warning threshold to prevent the vehicle battery temperature from rising sharply and becoming uncontrollable, leading to accidents.

[0083] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0084] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0085] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle, characterized in that: The method includes: Obtain the temperature data sequence and voltage data sequence of the battery when the car is running; obtain the lag order using the temperature data sequence, and obtain the temperature sequence to be analyzed according to the lag order and the temperature data sequence; The voltage data at the same time as each data of the temperature sequence to be analyzed is obtained in the voltage data sequence to form a voltage sequence; the state bit of each voltage data is obtained by using the difference between every two adjacent data in the voltage sequence; The state bits of each voltage data are combined into a state bit sequence, and the state bit sequence is segmented to obtain different state bit data segments; the state bit data segments are mapped to the temperature sequence to be analyzed to obtain temperature data segments; A window is established with a temperature data as the center in a temperature data segment, and the interference coefficient of the temperature data is calculated according to the variance of the difference between every two adjacent data in the window and the data before and after the temperature data; According to the difference and slope value between each two adjacent data in the temperature sequence to be analyzed, the intensity coefficient of each temperature data showing a sharp upward trend is obtained; according to the intensity coefficient and interference coefficient of the temperature data showing a sharp upward trend, the prediction weight of the temperature data is obtained; Based on the prediction weight of each temperature data in the temperature sequence to be analyzed, the ARIMA algorithm is used to obtain the predicted temperature at the time to be predicted; the vehicle battery temperature warning is performed according to the predicted temperature; The step of obtaining the prediction weight of the temperature data according to the intensity coefficient and the interference coefficient showing a sharp upward trend of the temperature data includes: Calculate the difference between the preset value and the interference coefficient of a temperature data and multiply it by the intensity coefficient of the sharp upward trend of the temperature data to obtain the credibility of the sharp upward trend of the temperature data; calculate the inverse of the time interval between the moment corresponding to the temperature data and the moment to be predicted, which is recorded as the degree of proximity; calculate the average value of the credibility of the sharp upward trend of the temperature data and the degree of proximity of the temperature data and normalize them to obtain the prediction weight of the temperature data.

2. The method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle according to claim 1, characterized in that: The step of obtaining a temperature data sequence and a voltage data sequence of a battery when the vehicle is running includes: The battery management system BMS is used to collect real-time temperature data and voltage data of the vehicle driving battery to form a temperature data sequence and a voltage data sequence respectively, and the collection time interval is a preset interval.

3. The method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle according to claim 1, characterized in that: The method of obtaining the lag order by using the temperature data sequence and obtaining the temperature sequence to be analyzed according to the lag order and the temperature data sequence comprises: The ACF and PACF diagrams are drawn using the collected temperature data sequence, from which the lag order P is obtained; according to the lag order, the last P temperature data in the temperature data sequence are selected to form the temperature sequence to be analyzed.

4. The method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle according to claim 1, characterized in that: The step of obtaining the state bit of each voltage data by using the difference between every two adjacent data in the voltage sequence comprises: The status bit is represented as 0 or 1, 0 represents the discharge state, and 1 represents the charge state. The difference between every two adjacent voltage data in the voltage sequence is calculated, that is, the latter voltage data minus the previous voltage data. If the difference is greater than or equal to zero, the status bit of the latter voltage data in the adjacent voltage data is 1. If the difference is less than zero, the status bit of the latter voltage data in the adjacent voltage data is 0.

5. The method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle according to claim 1, characterized in that: The step of segmenting the status bit sequence to obtain different status bit data segments includes: Traverse from the first state bit of the state bit sequence. If a data segment with the same state bits and the number of the same state bits is greater than or equal to the preset number is found during the traversal process, it is recorded as the starting segment. The traversal starts again from the next state bit of the starting segment. If a data segment with a state bit different from the starting segment is found, but the state bits are also the same and the number of the same state bits is greater than or equal to the preset number, it is recorded as the ending segment. The traversal stops, and the state bits between the starting segment and the previous data of the ending segment constitute the state bit data segment. Similarly, the traversal continues from the ending segment. At this time, the ending segment is the starting segment of the next state bit data segment, and the state bit data segments in the state bit sequence are obtained.

6. The method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle according to claim 1, characterized in that: The step of mapping the state bit data segment to the temperature sequence to be analyzed to obtain the temperature data segment includes: The position of each state bit data segment in the state bit sequence is mapped to the temperature sequence to be analyzed, the temperature sequence to be analyzed is segmented, and the temperature data segment corresponding to each state bit data segment is obtained.

7. The method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle according to claim 1, characterized in that: The calculation formula of the interference coefficient of the temperature data is: ， in, Represents the interference coefficient of the i-th temperature data in the temperature sequence to be analyzed; represents the variance of the difference between every two adjacent data in the window centered on the i-th temperature data in the temperature sequence to be analyzed; m represents the number of temperature data in the window; represents the variance of the difference between every two adjacent data in the m+ temperature data after the i-th temperature data in the temperature data segment to which the i-th temperature data belongs; represents the variance of the difference between every two adjacent data in the m-temperature data before the i-th temperature data in the temperature data segment to which the i-th temperature data belongs; It represents the distance between the i-th temperature data and a straight line fitted by using other data except the i-th temperature data in the temperature data segment to which the i-th temperature data belongs.

8. The method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle according to claim 1, characterized in that: The step of obtaining the intensity coefficient of each temperature data showing a sharp upward trend according to the difference and slope value between each two adjacent data in the temperature sequence to be analyzed includes: The difference between every two adjacent data in the temperature sequence to be analyzed is used to form a difference sequence, the difference is the subtraction of the previous data from the next data, and the difference between every two adjacent data is recorded as the difference corresponding to the next data; if the difference corresponding to a temperature data in the temperature sequence to be analyzed is less than or equal to zero, then the intensity coefficient of the temperature data showing a sharp upward trend is zero; if the difference corresponding to a temperature data in the temperature sequence to be analyzed is greater than zero, then the slope between the temperature data and the previous temperature data is taken as the slope term, the absolute value of the difference between the difference corresponding to the temperature data and the maximum value in the difference sequence is obtained, and the difference difference term is obtained by inverting it, and the average of the slope term and the difference difference term is the intensity coefficient of the temperature data showing a sharp upward trend.

9. The method for early warning of battery temperature of an on-board terminal of an intelligent network-connected vehicle according to claim 1, characterized in that: The vehicle battery temperature warning according to the predicted temperature includes: Set the battery temperature threshold. If the predicted temperature is greater than or equal to the battery temperature threshold, an early warning is issued.

Images