[0044] The preferred embodiments of the present invention will be described in detail below.
[0045] As shown in Figure 1, the present invention provides a power battery management system that integrates control and protection functions. It can be applied to electric vehicles. It includes a multifunctional detection board 100 for detecting high-voltage battery packs and their supply, The abnormal working signal of the power consumption system; it also includes: a battery control management module 200, a multi-parameter fuzzy processing module 320, and a controller circuit 400; the multi-parameter fuzzy processing module 320 is used to sequentially perform detection signals output by the multi-function detection board 100 Fuzzy processing, membership calculation, fuzzy inference classification and defuzzification processing process, and judge the current operating state of the automobile system (the automobile system includes the power battery management system of the present invention and the high-voltage battery pack and its power supply and power system); The battery control management module 200 is used to send the detection signal from the multi-function detection board 100 to the multi-parameter fuzzy processing module 320, and give control commands according to the judgment result of the multi-parameter fuzzy processing module 320; the controller circuit 400 is used to execute The control command issued by the battery control management module 200 protects the high-voltage battery pack and its power supply and consumption system. Among them, the battery control management module 200 and the multi-parameter fuzzy processing module 320 can be implemented by an MCU, and a part of the data storage and processing space of the MCU is divided and used for multi-parameter fuzzy processing. Then when the MCU receives the input from the multi-function detection board 100 When multiple detection signals are detected, they are directly stored in the corresponding data storage and processing space.
[0046] It can be seen that the present invention introduces a fuzzy inference method of intelligent decision-making method in the power battery management system, which can effectively deal with multi-input problems, and at the same time does not require accurate numerical judgment conditions, only through the setting of the membership function, Fuzzy reasoning can be carried out, and the system can integrate multiple possibilities into a maximum possible output through the process of defuzzification, and obtain a unique final judgment, realizing the mapping from multiple inputs to single output. Therefore, according to the fuzzy inference method, the power battery management system of the present invention summarizes the operating conditions of multiple automobile systems corresponding to multiple operating parameters into four conditions: normal operation, slight deviation, deviation alarm and action protection four working modes. When the automobile system is in normal operation and slight deviation mode, the power battery management system of the present invention does not perform any protection action; when the automobile system is in the deviation alarm mode, the power battery management system of the present invention starts an alarm or/and gives a fault prompt ; When the car system is in the action protection mode, the power battery management system of the present invention starts the protection mode; see Figure 2, the protection mode mentioned here specifically refers to:
[0047] (1) If the fault occurs in the normal working stage, the positive and negative poles of the battery bus are disconnected through the main feed control relays 1 and 2 to achieve complete isolation between the battery and the external circuit, and the charge discharge switch 4 is turned on for a certain period of time. Reset after the residual voltage on the line drops to within the safe limit. If the fault occurs in the pre-charging stage, disconnect the positive and negative poles of the battery bus through the main feed control relay K1, K2 and the pre-charge control relay K3 to achieve complete isolation between the battery and the external circuit, and turn on the charge discharge switch K6 for a certain amount Time to reset the residual voltage on the line after it drops to a safe limit. The main feed control relay K1 controls the opening and closing of the negative pole of the battery bus, the main feed control relay 2 controls the opening and closing of the positive pole of the battery bus, the pre-charge control relay K3 controls the opening and closing of the pre-charging branch, and the charge discharge relay switch K6 controls The above-mentioned relays are controlled by the controller circuit 400 to open and close the charge discharge circuit.
[0048] (2) Identify and code the type of failure.
[0049] (3) Send the fault code to the upper-level controller (that is, the vehicle controller). The battery control management module 200 of the system of the present invention can be offline or online to insert new programming data, which can be the same as the vehicle controller in the vehicle or other The control device communicates to exchange processing information. In addition to receiving instructions from the upper-level controller, it can also transmit local control information, such as fault codes, etc., upwards. Its communication can use RS232 serial port or CAN bus.
[0050] (4) The fault display, draw the driver's attention.
[0051] 4, the multi-parameter fuzzy processing module 320 includes: at least one fuzzification unit 321, at least one membership degree calculation unit 322, a fuzzy inference rule base 323, and a defuzzification unit 324; each detection signal output by the multi-function detection board 100 The detection signal is fuzzified by a fuzzification unit 321, and a fuzzy signal is obtained; each fuzzy signal is classified and processed by a membership calculation unit 322; the fuzzy inference rule library 323 is based on the result of the above membership calculation , Derive the probability that the current automobile system operating state belongs to various situations; the defuzzification unit 324 maps the multiple possibilities of the automobile system operating state to a result according to the probability of the current automobile system operating state belonging to various situations, And give the only operating state mode of the current car system. It can be seen from Fig. 4 that after multiple parameters such as insulation signals are input to the multi-parameter fuzzy processing module 320, the current operating state of the automobile system is finally determined, namely, among the four working modes of normal operation, slight deviation, deviation alarm and action protection. Kind of. The principle is as follows:
[0052] As shown in Figure 4, the present invention can obtain vehicle monitoring points: leakage signal, overcurrent signal, insulation signal, battery circuit terminal voltage signal, load voltage signal, 4 sets of battery branch detection signals, and 2 sets of temperature detection signals , A total of 11 sets of signals. These signals are continuous quantities that can be quantified. According to the principle of fuzzy inference, each group of signals is regarded as an input variable. For each input variable, four lingual variables (Lingual Variables) of no fault, critical fault, minor fault, and serious fault are selected. Each semantic variable sets the membership function (MembershipFunction). The reasoning system uses an output variable, that is, the operating conditions of the system. This variable is divided into four language variables: normal, deviation, alarm, and protection. The reasoning process adopts the standard Mamdani reasoning method. The block diagram of the multi-parameter fuzzy processing module is shown in Figure 4, which has the following aspects.
[0053] 1. The collection of multiple parameters can be obtained from the detection circuit and transmitted to the MCU, or transmitted through the CAN bus. The fuzzification method adopts the typical single-point fuzzy (Singleton Fuzzification) method.
[0054] 2. Setting of membership function. The membership function expresses the classification of input variables, namely linguistic variables (a linguistic variable is a category). Given an input quantity, the probability of the input quantity belonging to a certain language variable and the degree of membership can be obtained through the calculation of the membership function.
[0055] Regarding the setting of the membership function, the leakage detection signal is taken as an example to illustrate: the system of the present invention requires the leakage current to be less than 30 mA to ensure the safety of the human body and the stability of the electrical system of the car. Therefore, the four language variables of the leakage signal are defined as (unit mA):
[0056] No trouble
[0057] According to the actual situation and the judgment of the severity of the fault, the present invention artificially selects the triangular function and its related parameters as the membership function of the four language variables, as shown in FIG. 5. According to the above method, four language variables and their corresponding membership functions are set for each detection signal.
[0058] 3. Inference rule library setting. Inference rules establish a logical connection between input variables and output variables. Given a set of combinations of input variables, an output situation can be derived. Therefore, it is necessary to set up an inference rule library for the fuzzy inference system, which is composed of several "if...then..." conditional judgment sentences. The content of the If guide is the combination of the input variables, and the content of the then guide is the output caused by the combination of the above situations. E.g:
[0059] If leakage signal = serious fault & over current signal = minor fault & load voltage = minor fault then car system operation status = serious fault
[0060] 4. Fuzzy reasoning method. Fuzzy reasoning is the core step of the fuzzy decision-making system. It can deduce the probability that the current automobile system operating state belongs to various situations based on the calculation results of the membership degree of each input quantity and based on the above-mentioned fuzzy rule library. The invention adopts Mamdani inference rules to perform fuzzy inference. The Mamdani reasoning process is a standard fuzzy reasoning process, which belongs to the existing technology and will not be described in detail here.
[0061] 5. Method of defuzzification. Based on the probability that the operating state of the automobile system belongs to various situations given by the fuzzy inference process, the present invention adopts the typical center of gravity method to realize the fuzzification process, and maps multiple possibilities into a final conclusion. The only judgment for the current operating state of the automotive system. According to the above method and through the standard Mamdani reasoning process, the present invention can comprehensively and fuzzy map 11 sets of detected signals into 4 kinds of automobile system operation conditions, namely normal operation, slight deviation, deviation alarm and action protection four working modes. And through the battery control management module 200, the four situations and the corresponding faults are handled respectively.
[0062] In addition, in order to adapt to different vehicles and different applications, the membership function and trigger threshold of each fault language variable can be set online. The online setting method is as follows: the upper controller (such as the vehicle controller) communicates with the system through the CAN bus, and the parameters of each criterion and threshold are issued or modified through the write operation, and the current judgements are read in through the read operation. Data and threshold parameters. After the criterion and threshold parameters in the system software are modified, it will run in real time according to the new parameters.
[0063] As shown in Figure 1, the multifunctional detection board 100 includes: an insulation measurement and control module 110, a leakage detection module 120, an overcurrent detection module 130, a battery loop terminal voltage detection module 140, a load voltage detection module 150, a temperature shunt detection module 160, and Power battery branch detection module 170.
[0064]Referring to Figure 2, the insulation measurement and control module 110 is used to test the insulation resistance between the high-voltage battery pack 111 and its feeder line, and sends the insulation signal to the battery control management module 200; the insulation measurement and control module 110 can also be used to detect the main circuit The principle of equivalent insulation resistance of its related equipment is shown in Figure 3: RJ+ and RJ- are measured according to the method described in GB/T-18384.1-1-2001. Due to the special environmental factors in the car, the high-voltage circuit is insulated Resistance is easy to constitute a dynamically changing physical parameter, and its size is related to the number of high-voltage electrical appliances in the high-voltage circuit loop and the state of electricity consumption. For this reason, the static detection and online dynamic monitoring of the high-voltage electrical insulation status are the key to safety diagnosis. It integrates the insulation status between the battery pack and its feeder circuit, high-voltage electrical devices, motor drive systems and other electrical appliances and the car body ,As shown in Figure 3. Use the additional known resistors R4 and R5 to attenuate the voltage signal partial pressure to dynamically measure the equivalent insulation resistance RJ+ and RJ- of the positive and negative output terminals of the battery pack relative to the body; Ri1 and Ri2 are the equivalent internal resistance of the battery pack When calculating equivalent insulation in high voltage circuit, its value is so small that it can be ignored. Among them, V1 and V2 are the voltages of the positive and negative terminals of the battery to the ground before K4 and K5 are closed, respectively, and the voltages after K4 and K5 are closed are V1` and V2` respectively. The opening and closing of K4 and K5 are automatically operated by the system through the control relay. The values of equivalent insulation resistance RJ+ and RJ- are as follows:
[0065] RJ+=R4(1+V2/V1)(V1-V1`)/V1`
[0066] RJ-=R5(1+V1/V2)(V2-V2`)/V2`
[0067] Considering that the limit of insulation resistance has a great relationship with the withstand voltage of high-voltage electrical appliances, the equivalent insulation resistance RJ+ and RJ- are calculated and divided by the current battery pack voltage (V1+V2), according to 100-500 ohms/ Voltage standard, distinguish and diagnose the fault level of insulation resistance.
[0068] When an insulation fault occurs, start the fault diagnosis program and determine whether the fault is gradual or sudden by judging the trend of the fault. If the fault occurs suddenly and its value is far below the operating standard, the protection mode will be quickly activated (specific instructions will be given below), and the upper-level controller (ie, the vehicle controller) will be notified with a CAN event frame. The upper-level controller shall Feedback the corresponding processing results.
[0069] 2, the leakage detection module 120 is used to detect the leakage of the feeder circuit, the traction motor 121 and the inverter 122, and send the leakage signal m to the battery control management module 200, which uses a dedicated Hall leakage The sensor LD, when the leakage current reaches 10-30mA and the duration is more than 20mS, the above protection mode is quickly activated (the protection threshold is adjustable).
[0070] Referring to FIG. 2, the overcurrent detection module 130 is used to detect the current values of the two overcurrent sensors GL1 and GL2 on the feeder line of the high-voltage battery pack 111, and output different overcurrent signals e and f according to the detected results. Battery control management module 200.
[0071] Referring to FIG. 2, the battery loop terminal voltage detection module 140 is used to detect the voltage value at the end of the high-voltage battery pack 111, that is, the voltage VB+ at point B+ and the voltage VB- at point B-, and the voltage signals representing the values of VB+ and VB- a, b are transmitted to the battery control management module 200.
[0072] 2, the load voltage detection module 150 is used to detect the voltage value of the feeder line of the high-voltage battery 111 under the load state, that is, the voltage VH+ at the H+ point and the voltage VH- at the H- point, and the voltages representing the values of VH+ and VH- The signals c and d are transmitted to the battery control management module 200.
[0073] Referring to FIG. 2, the temperature branch detection module 160 is used to detect the temperature values of multiple battery groups inside the high-voltage battery pack 111, and transmit the temperature signal to the battery control management module 200, which can be placed on the high-voltage battery pack 111 through monitoring Figure 2 and Figure 1 only show two temperature sensors T1 and T2. In fact, there can be more than two temperature sensors. The temperature sensor is determined by the battery grouping inside the high-voltage battery 111.
[0074] Referring to FIG. 2, the power battery branch detection module 170 is used to detect the branch voltages of multiple battery groups in the high-voltage battery pack 111, and send the multiple voltage division signals to the battery control management module 200. The battery control management module 200 The multi-channel voltage division value calculates the total voltage value and the remaining power of the high-voltage battery pack 111, and is used to determine the current operating condition of the high-voltage battery pack 111 according to the total voltage value. How to detect the shunt voltages of multiple battery groups in the high-voltage battery pack 111 is common knowledge and will not be described in detail here.
[0075] As shown in FIG. 1, in the system of the present invention, in addition to a multi-parameter fuzzy processing module 320 that performs intelligent fuzzy judgment on multiple parameters, there is also a quick-decision processing module 310, which is used to receive the output of the multi-function detection board 100. The detection signal (that is, the above-mentioned insulation measurement and control module 110, leakage detection module 120, overcurrent detection module 130, battery circuit terminal voltage detection module 140, load voltage detection module 150, temperature branch detection module 160, and power battery branch detection module 170 Output detection signals of multiple operating parameters), and compare the corresponding danger thresholds of the detection signals of each parameter. When one of the detection signals exceeds the danger threshold, the module sends a command to the controller circuit 400 to start the protection mode. The present invention adds a quick decision processing module 310 and serves as an auxiliary device of the multi-parameter fuzzy processing module 320, thereby improving the rate and accuracy of the system in judging system failures caused by a single operating parameter exceeding the standard.
[0076] As shown in Figure 1, in addition to the above components, the system of the present invention also includes: a manual disconnection emergency treatment detection module 420, which is used to detect the closing or conduction of the manual emergency button set in the automobile system. Once the button is manually activated, The manual disconnection emergency processing detection module 420 sends the signal that characterizes the emergency disconnection (for example, setting the manual emergency button to turn on to characterize the emergency disconnection) into the battery control management module, and when the battery control management module confirms that the signal exists, it sends the signal to the The controller circuit issues a command to activate the protection mode. Due to the important status of this button, this button should generally have more than two control points to prevent misoperation.
[0077] As shown in Fig. 1, the system of the present invention also includes: a multifunctional interlock detection module 430, which is used to detect the interlock status signal sent by the host controller or other controllers, and transmit the signal to the battery control management module 200, When the battery control management module 200 determines that the vehicle control device has an interlock request, the battery control management module sends out an alarm signal or a fault prompt message. This is mainly embodied that the system of the present invention is combined with the existing multifunctional interlocking device of the automobile to provide a fault judgment signal for the system. The multifunctional interlocking detection module 430 is a common device in the existing automobile, and will not be described in detail here.
[0078] As shown in FIG. 1, the system of the present invention may further include: a passive safety control module 450, which is used to detect the signal of the acceleration sensor arranged in the control loop of the vehicle control device, and send the signal to the battery control management module When the battery control management module 200 determines that the signal exceeds the limit, the battery control management module 200 sends a command to the controller circuit to activate the protection mode. Similarly, a passive safety control trigger module 180 can be provided on the multifunctional detection board 100, which is used to detect the signal of an acceleration sensor set in the system control loop, and send the signal to the quick decision processing module 310 for thresholding. For comparison, specifically, a detection quantity can be derived from the output of the passive safety control module 450 as the detection signal of the passive safety control trigger module 180. The purpose of the passive safety control module 450 and the passive safety control trigger module 180 is to detect abnormal working conditions of the automobile system, that is, when the vehicle collides with the vehicle and causes the high-voltage pile head to loosen, the high-voltage main circuit short circuit and other abnormal conditions, it needs to be automatically Interlock and quickly activate the protection mode. The detection of abnormal conditions can be implemented by setting one to several acceleration sensors. When a vehicle collides, an acceleration signal exceeding the conventional one is generated, which is collected by the acceleration sensor and sent to the MCU. The MCU determines whether the signal exceeds the standard, and adopts a protection mode if it exceeds the standard.
[0079] The above is a description of the system structure. Correspondingly, the present invention also provides a power battery management method that integrates control and protection functions. The method is performed according to the following steps:
[0080] A. The above-mentioned power battery management system is initialized;
[0081] B. Detect various working parameters of the high-voltage battery pack and abnormal working signals of the vehicle control device to obtain detection signals;
[0082] C. Fuzzy processing, membership calculation, fuzzy inference classification and defuzzification processing are sequentially performed on the detection signal to obtain the operating result of the unique operating state mode of the current automobile system;
[0083] D. The battery control management module gives corresponding control commands according to the operation result, that is, when the multi-parameter fuzzy processing module 320 determines that the current operating state of the automobile system belongs to the action protection mode, the battery control management module 200 sends to the controller circuit 400 The command to start the protection mode, the command includes the main feed control 1 command to control the action of the main feed control relay K1, the main feed control 2 command to control the action of the main feed control relay K2, and the pre-charge control relay K3 to control the action. The power-on control 3 command, the charge discharge switch 4 command that controls the action of the charge discharge switch K6; otherwise, the step A4 is executed. For how to execute it, please refer to the flow shown in FIG. 7; when the multi-parameter fuzzy processing module 320 determines the current operating state of the automobile system When it belongs to the deviation alarm mode, the battery control management module 200 sends out an alarm signal or a fault prompt message.
[0084] As shown in Figure 6, the above initialization process may include the following steps:
[0085] A1, initialize the operating system;
[0086] A2. The battery control management module 200 determines whether the system is working normally; if yes, execute step A3; otherwise, send a command to start the protection mode to the controller circuit 400;
[0087] A3. Scan the working status of the passive safety control module 440 and the multi-function interlocking module 430, and determine whether the signals output by the passive safety control module 440 and the multi-function interlocking module 430 exceed the standard; if yes, the battery control management module 200 sends to the controller The circuit 400 issues a command to start the protection mode;
[0088] A4. Start the system communication function, and determine whether the command to manage the power battery from the vehicle controller is received; if yes, perform step A5; otherwise, the battery control management module 200 sends a command to the controller circuit 400 to start the protection mode; The judgment here involves judging the correctness of the upper command. For example, if the CAN communication method is used, if the CAN execution command is received incorrectly, it also indicates a system failure;
[0089] A5. Start the controller circuit 400, the multi-function detection board 100, and the multi-parameter fuzzy processing module 320.
[0090] Referring to the above system description, the fuzzification processing in step C adopts the single-point fuzzy method, and the triangle function and its related parameters are selected as the membership function of the language variable in the membership calculation process.
[0091]As shown in Fig. 6, between step B and step C of the method, the following quick decision step can be added: the detection signal output by the multifunctional detection board 100 is sent to the quick decision processing module 310, and the quick decision processing module 310 separately evaluates the parameters The detection signal is compared with the threshold value. When one of the detection signals exceeds the dangerous threshold value, the module sends a command to the controller circuit to start the protection mode. In the quick decision step, when the detection signal input by the temperature branch detection module 160 to the quick decision processing module 310 exceeds the dangerous threshold, the quick decision processing module 310 sends a temperature control instruction to the controller circuit (as shown in the temperature control 1 in Figure 1). And 2 can be used to control the fan of the battery pack cooling and cooling system) to control the battery pack cooling and cooling system.
[0092] For electronic control devices, the environmental working conditions of automobiles are quite harsh. This environment has strong shocks and vibrations, high temperature flashes, corrosive gases, strong electromagnetic radiation interference, and so on. Therefore, a reasonable integrated design is obviously extremely important. The integrated measurement, control and protection module (all functional modules shown in Figure 1) is placed in a sealed metal cavity shell, and the circuit port is connected to each sensing and control device through a sealed aerial plug to supply power and sensors. Multiple interference shielding and suppression measures are used in links such as the layout of the controller and the controller. For example, as shown in FIG.
[0093] The rapid decision processing and multi-parameter fuzzy processing compound control method adopted in this scheme, as well as the proposed protection mode, have good real-time control and multi-parameter optimization processing effects, and bring together most of the most important requirements for electric vehicles and hybrid vehicles. The measurement and control function, with its promotion and application, will play a positive role in promoting the development of electric vehicles and hybrid vehicle projects to a certain extent.
[0094] It should be understood that those of ordinary skill in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the protection scope of the appended claims of the present invention.
