High-robustness daisy chain breakage detection method and system

Abstract

The invention discloses a daisy chain breakage detection system with high robustness, and belongs to the technical field of vehicle-mounted software, the system comprises a power supply logic circuit module, a power supply chip, a master control module, a communication isolation module and a plurality of slave control units, the isolation module is provided with more than two paths of SPI communication interfaces, and each slave control module is characterized by comprising a plurality of slave control units. Each slave control unit is composed of N AFEs, parameter information is transmitted between the slave control units and the master control module and between the AFEs in the slave control units through daisy chain communication, master control instructions and configuration information are received, in the dormant state of the system, the AFEs periodically generate heartbeat signals, the communication state of the daisy chain is reflected in real time, and the communication time of the daisy chain is shortened. The number of abnormal chain breakage and the specific chain breakage position in the daisy chain communication process can be quickly and accurately locked through a software means, and an additional hardware circuit is not needed.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle-mounted software, and in particular relates to a daisy chain break detection method and system with high robustness. Background Art

[0002] The battery management system BMS mainly realizes important functions such as information collection, status diagnosis, high-voltage monitoring, relay control, and vehicle communication of lithium-ion batteries. The BMS system usually includes a master control BMU and multiple slave control CMUs. As the demand for long-range driving of the vehicle becomes higher and higher, the number and capacity of battery cells are also increasing. The number of CMU monitoring units also increases with the increase in the number of battery cells. In this way, more front-end analog sampling chips AFE (Analog Front End analog front-end chips) are needed to monitor the status of each battery cell in real time. Daisy chain communication is mainly used between BMU and slave control CMU and between adjacent front-end analog sampling chips AFE in CMU. All battery information including battery cell voltage, temperature, balance status, etc. must be transmitted to BMU through the daisy chain communication link to enable BMU to monitor and judge the battery cell status and ensure system safety. Compared with traditional high-cost CAN communication, the anti-interference ability and connection stability of daisy chain communication are greatly weakened. When the daisy chain communication is interfered or a chain break occurs, the BMU will not be able to obtain battery information, and all battery cells will be out of connection and out of control, and safety cannot be guaranteed. This requires the BMU to be able to make quick and accurate judgments on daisy chain communication anomalies so that the entire vehicle can respond to corresponding faults and ensure system safety.

[0003] Currently, the diagnosis of daisy chain communication mostly relies on whether the CRC check of AFE has errors. The check logic is relatively simple and it is easy to misdiagnose the interference as a chain break. It has poor robustness and cannot accurately determine whether it is a single-point fault or a multi-point fault, nor can it accurately diagnose the specific location of the chain break. Summary of the invention

[0004] The purpose of the present invention is to solve the problems in the prior art that the robustness of daisy chain communication break diagnosis is poor, and the break position and the number of break are difficult to determine, thereby providing a complete set of software logic and system solutions. Under this system solution, it is possible to quickly and accurately lock whether the break mode is a single-point break or a multi-point break and the specific break position, greatly improving the robustness of daisy chain diagnosis. Furthermore, the system solution can realize two-way communication of the daisy chain, and can still realize real-time detection of all battery cells in a single-point break fault mode. Even in the system sleep state, the system solution can realize real-time detection of battery cells, and can actively reverse wake up the BMS system when an abnormal state is detected, so as to achieve the ability of active warning of thermal runaway.

[0005] To achieve the above-mentioned purpose, the technical solution of the present invention is as follows: a highly robust daisy chain break detection system, characterized in that the system includes a power supply logic circuit module, a power chip, a master control module, a communication isolation module and a plurality of slave control units, the isolation module has more than two SPI communication interfaces, the slave control module is characterized in that it is composed of multiple slave control units, each slave control unit is composed of N AFEs, and the slave control units and the master control modules and the AFEs in the slave control units transmit these parameter information and receive master control instructions and configuration information through daisy chain communication. When the system is in sleep mode, the AFE periodically generates a heartbeat signal to reflect the daisy chain communication status in real time.

[0006] As an improvement of the present invention, a daisy chain break detection method with high robustness includes: The chain break software diagnosis strategy during the system initialization process detects the chain break fault and chain break position during the initialization phase. The system will continue to monitor the battery cell status information and continue to try to restore the daisy chain communication. When a suspected chain break is detected during the initialization phase and a reset is attempted, all fault interfaces of the AFE are cleared. The method includes a link-break software diagnosis strategy during the normal operation of the system, wherein the link-break diagnosis strategy during the system operation phase is triggered by PEC (Packet Error Code) fault and data invalid fault events; During the diagnosis process in the operation phase, after the multi-point link break fault is confirmed, the system continues to collect bidirectional data and determines whether the bidirectional data combination is complete. If the data is complete, the multi-point fault state is exited; otherwise, the multi-point fault state is maintained.

[0007] As an improvement of the present invention, the initialization phase link break diagnosis strategy steps include: S1, in the initial stage, the daisy chain bidirectional function check is first performed to determine whether the data has PEC anomalies and invalid data. If there are no anomalies, there is no chain break fault; when the collected data has PEC error & node, the slave control unit CMU data is invalid or the end chain break detection is abnormal, continuous reading is performed; S2, read once in the forward direction and once in the reverse direction, which is counted as one cycle; S3, judging whether the read data is normal and the disconnection information of the erroneous data; S4, execute the above sequence S1 to S3 in a loop, and record the number of loops N; S5, determining whether the number of consecutive cycles N ≥ 3; S6, if N≥3, perform a reset operation, reinitialize, and repeat S1 to S4; S7: If N≥3 is still satisfied, a multi-point fault is reported, and the corresponding broken link position is reported.

[0008] As an improvement of the present invention, the S3 judgment process includes: S31, if normal data is read unidirectionally during the loop, the loop is exited and a no-broken-link fault is reported; S32: If the unidirectional data are all wrong and the bidirectional data combination is complete for 10 consecutive cycles, determine whether the error is at the same position. If it is at the same position, determine that the single point is broken and report the broken position, and jump to bidirectional communication. If it is not at the same position, report the single point fault and jump to bidirectional communication. The broken position report is uncertain. If the single point fault disappears during the bidirectional communication, jump to S31. If the bidirectional combination data is incomplete during the bidirectional communication, jump to S33 for judgment. S33, if there are errors in the unidirectional data and the combination of the bidirectional data is incomplete for 10 consecutive cycles, proceed to step S4.

[0009] As an improvement of the present invention, the chain break diagnosis strategy steps in the system operation phase are as follows: S1, if there is no PEC problem in data transmission and the data is valid, the link break detection will not be triggered; when the collected data has a PEC error and the data received by the slave control unit after the node is invalid, the data validity of the data segment is immediately set to invalid, and the link break detection is started, and continuous data reading begins; S2, read once in the forward direction and once in the reverse direction, which is recorded as one cycle; if the one-way communication data returns to normal during the cycle, the data validity is set to valid; if the data combination is normal during the cycle, the data validity is immediately set to valid; if the forward and reverse data combination is incomplete, the lost data is set to invalid, and the readable data is uploaded normally and the data is valid; S3, judging whether the read data is normal and the disconnection information of the erroneous data; S4, then execute the above sequence S1 to S3 in a loop and record the number of loops N; S5, determining whether the number of consecutive cycles N ≥ 3; S6, if N ≥ 3, perform a reset operation to reinitialize; and repeat S1 to S4; S7: If N≥3 is satisfied for three consecutive cycles, multiple link breaks are determined and the link break locations are reported.

[0010] As an improvement of the present invention, the determination step in S3 includes: S31, if normal data is read unidirectionally during the loop, the loop is exited; S32: If the unidirectional data are all wrong and the bidirectional data combination is complete for 10 consecutive cycles, determine whether the error is at the same position. If it is at the same position, determine that the single point is broken and report the broken position, and jump to bidirectional communication. If it is not at the same position, report the single point fault and jump to bidirectional communication. The broken position report is uncertain. If the single point fault disappears during the bidirectional communication, jump to S31. If the bidirectional combination data is incomplete during the bidirectional communication, jump to S33 for judgment. S33: If there are errors in the unidirectional data and the combination of the bidirectional data is incomplete for 10 consecutive cycles, then proceed to S4.

[0011] As an improvement of the present invention, in order to implement the monitoring of the cell status, all fault interfaces of the AFE are cleared when a reset attempt is made to recover.

[0012] As an improvement of the present invention, during the process of detecting and confirming multi-point link breakage, if the two-way communication data combination is incomplete, the lost data is set to be invalid, and the readable data is uploaded normally and the data is valid.

[0013] As an improvement of the present invention, when a reset attempt is made to recover, all collected data retain the last value, and the data validity is set to invalid until the reset is recovered, and then the sampled data is updated.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1) The number and specific location of abnormal chain breaks during daisy chain communication can be quickly and accurately locked by software, without the need for additional hardware circuits; 2) It has the characteristics of adaptive selection of diagnostic logic and can execute different diagnostic logic according to the working status of the system; 3) It has the capability of two-way communication, and its characteristic is that it can still monitor all battery cell parameter information in the event of a single point of disconnection; 4) It has the ability to reverse wake up the system under abnormal conditions, and its characteristics can detect possible thermal runaway risks in a timely manner; 5) It has the ability to detect daisy chain communication anomalies and battery cell monitoring in an all-weather, real-time and adaptive manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a system architecture diagram of the present invention; Figure 2 This is a flow chart of the chain break diagnosis in the initialization phase of the present invention; Figure 3 Flow chart of the chain break diagnosis in the normal operation stage of the present invention; Figure 4 This is a schematic diagram of the reverse wake-up and two-way communication functions of the present invention; Figure 5 This is a reverse control flow chart of data transmission of the present invention; Figure 6 This is the flow chart of the slave control low power consumption working mode. DETAILED DESCRIPTION

[0016] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0017] Embodiment 1: Figure 1 As shown, Figure 1 The architecture of the entire system solution is demonstrated, which mainly includes a power supply logic circuit module LDO, a power chip SBC, a master control BMU, an SPI communication isolation module ISOSPI, and several slave control units CMU*n (for ease of understanding, the present invention takes n=4 as an example). The power supply characteristics of LDO and SBC are always powered (i.e., the system can enter the dormant state of low power consumption mode); the characteristic of the BMU module is the brain of the system, running software programs, processing data and responding, and controlling the working state of the entire system. The diagnostic logic involved in the present invention is also running in this module; the characteristic of the isolation module ISOSPI is that it has more than two SPI communication interfaces, namely Port1 and Port2, which are the bridges for slave control and master control data transmission, and play the functions of high and low voltage isolation, two-way communication, and releasing the interrupt INTR hard-wire signal to reversely wake up the BMS system when abnormal data is detected; the characteristic of the slave control module is that it is composed of multiple CMUs, each CMU is composed of N AFEs (for ease of understanding, the present invention takes N=3 as an example), which is responsible for monitoring all cell voltage, temperature, balance and other parameter information, and the CMU and BMU and the AFEs in the CMU transmit these parameter information and receive master control instructions and configuration information through daisy chain communication. In the dormant state of the system, the AFE periodically generates a "heartbeat signal" to reflect the daisy chain communication status in real time.

[0018] Chain break diagnosis strategy during initialization A highly robust daisy chain break detection scheme and system, which only needs to execute the break diagnosis strategy once during initialization; A highly robust daisy chain break detection scheme and system, the initialization phase break diagnosis process is as follows Figure 2 , the strategy is described as follows: Step 1: In the initial stage, the daisy chain bidirectional function is checked, that is, whether the forward reading data and the reverse reading data are abnormal; determine whether the data has PEC abnormalities and invalid data. If there are no abnormalities, there is no chain break fault; when the collected data has PEC error & invalid CMU data after the node OR abnormal end chain break detection, perform continuous reading; Step 2: Read once in the forward direction and once in the reverse direction, which is counted as one cycle; Step3. Judgment: A: If normal data is read unidirectionally during the loop, the loop will be exited and no link break fault will be reported; B: If the unidirectional data is all wrong and the bidirectional data combination is complete for 10 consecutive cycles, it is determined whether the error is in the same position. If it is in the same position, a single point fault is reported and the specific link break position is reported. If it is not in the same position, a single point fault is also reported and the link break position is reported as uncertain.

[0019] C: If there are errors in the unidirectional data and the bidirectional data combination is incomplete (multiple point breakages) for 10 consecutive cycles, proceed to Step 4 Step 4, then execute in a loop according to the above order Step 1-Step 3, and record the number of loops N; Step 5. Determine whether the number of continuous cycles N ≥ 3; Step 6. If N is greater than or equal to 3, perform a reset operation to reinitialize the system and repeat Step 1 to Step 4. Step 7. If N ≥ 3 is still satisfied, then report multiple point faults and the corresponding broken link locations; Example 2: Diagnostic strategy for interruption of normal system operation A highly robust daisy chain break detection scheme and system, in which a break detection mechanism is triggered by PEC failure and data invalid failure events during normal system operation; A highly robust daisy chain break detection scheme and system, the break diagnosis process in the normal operation stage of the system is as follows Figure 3 , the strategy is described as follows: Step 1: If there is no PEC problem in data transmission and the data is valid, the link break detection will not be triggered; if there is a PEC error in the collected data and the data received by the CMU after the node is invalid, the validity of the data segment is immediately set to invalid, and the link break detection is started, and continuous data reading begins; Step 2: Read once in the forward direction and once in the reverse direction, which is recorded as one cycle. If the one-way communication data returns to normal during the cycle, the data validity is set to valid. If the data combination is normal during the cycle, the data validity is immediately set to valid. If the forward and reverse data combination is incomplete, the lost data is set to invalid, and the readable data is uploaded normally and the data is valid. Step3. Judgment: A: If normal data is read unidirectionally during the loop, the loop will be exited; B: If the unidirectional data is wrong and the bidirectional data combination is complete for 10 consecutive cycles, determine whether the error is in the same position. If it is in the same position, determine that the single point is broken and report the broken position, and jump to bidirectional communication. If it is not in the same position, also report the single point fault and jump to bidirectional communication, and the broken position report is uncertain. If the single point fault disappears during the bidirectional communication process, jump to A. If the bidirectional combination data is incomplete during the bidirectional communication process, jump to state C for judgment; C: If there are errors in the unidirectional data and the bidirectional data combination is incomplete (suspected multi-point chain break) for 10 consecutive cycles, proceed to Step 4 Step 4, then execute in a loop according to the above order Step 1-Step 3, and record the number of loops N; Step 5. Determine whether the number of continuous cycles N ≥ 3; Step 6. If N ≥ 3, perform a reset operation to reinitialize, and repeat Step 1-Step 4. In order to monitor the status of the battery, in particular, 1) When the reset attempts to recover, all fault interfaces of AFE are cleared; 2) During the process of detecting multiple point disconnection confirmation, if the two-way communication data combination is incomplete, the lost data is set to invalid, and the readable data is uploaded normally and the data is valid; 3) When the reset is attempted, all collected data retains the last value, and the data validity is set to invalid until the reset is restored, and then the sampling data is updated.

[0020] Step 7. If N≥3 is satisfied for three consecutive cycles, multiple link breaks are determined and the link break locations are reported; In particular, after the multi-point disconnection fault is confirmed, the system continues to collect bidirectional data and determines whether the bidirectional data combination is complete. If the data is complete, the multi-point fault state is jumped; otherwise, the multi-point fault state is maintained; Example 3: Daisy chain two-way communication like Figure 4 , Figure 5 A highly robust daisy chain break detection scheme and system, wherein the ISOSPI IC has at least two SPI communication resources, Port1 and Port2, for forward data transmission and reverse data transmission respectively; A highly robust daisy chain break detection scheme and system, which transmits data in the forward direction when no communication anomaly is detected; A highly robust daisy chain break detection scheme and system, which switches to forward and reverse simultaneous data transmission for two-way communication when data anomalies or a break failure are detected; A highly robust daisy chain break detection scheme and system, daisy chain communication is restored to normal, and forward transmission data is switched back to improve communication efficiency; Example 4: Reverse wake-up of the system in sleep mode like Figure 4 , Figure 6 A highly robust daisy chain break detection scheme and system, after receiving a power-off command, will configure the slave control low power mode and reverse wake-up parameter threshold, and enter the sleep state after confirming the configuration result; In particular, the low power mode and reverse wake-up parameter thresholds are calibration quantities that can be calibrated; A highly robust daisy chain break detection solution and system. When the system is in sleep mode, the AFE will be in a low-power mode of cruise monitoring. In the cruise monitoring mode, the battery status information is detected in real time and a "heartbeat signal" is generated periodically. Once a threshold violation is detected, the abnormal "heartbeat signal" will be immediately transmitted to the ISOSPI IC. A highly robust daisy chain break detection solution and system, ISOSPI IC is always powered, and can still receive the "heartbeat signal" periodically sent by AFE in the system sleep state. When the "heartbeat signal" is abnormal, ISOSPI IC will release the INTR interrupt hard line signal, reversely wake up the BMS system for corresponding processing; In particular, the ISOSPI IC can monitor the cycle of the "heartbeat signal". Once the "heartbeat signal" is lost or the cycle does not meet expectations, the ISOSPI IC will also release the INTR interrupt hard-line signal to reversely wake up the BMS system for corresponding processing.

[0021] It should be noted that the above content only illustrates the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. For ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications all fall within the protection scope of the claims of the present invention.

Claims

1. A highly robust daisy chain break detection system, characterized in that: The system includes a power supply logic circuit module, a power chip, a BMU, a communication isolation module and several CMUs. The isolation module has more than two SPI communication interfaces. The slave control module is characterized by being composed of multiple CMUs, each CMU is composed of N AFEs. The CMU and BMU and the AFEs in the CMU transmit these parameter information and receive the main control instructions and configuration information through daisy chain communication. When the system is in sleep mode, the AFE periodically generates a heartbeat signal to reflect the daisy chain communication status in real time.

2. A highly robust daisy chain break detection method, characterized in that: Methods include: The chain break software diagnosis strategy during the system initialization process detects the chain break fault and chain break position during the initialization phase. The system will continue to monitor the battery cell status information and continue to try to restore the daisy chain communication. When a suspected chain break is detected during the initialization phase and a reset is attempted, all fault interfaces of the AFE are cleared. The method includes a software diagnostic strategy for link breakage during the normal operation of the system, wherein the link breakage diagnostic strategy during the system operation phase is triggered by PEC failure and data invalid failure events; During the diagnosis process in the operation phase, after the multi-point link break fault is confirmed, the system continues to collect bidirectional data and determines whether the bidirectional data combination is complete. If the data is complete, the multi-point fault state is exited; otherwise, the multi-point fault state is maintained.

3. A highly robust daisy chain break detection method according to claim 2, characterized in that: The initialization phase of the link break diagnosis strategy includes: S1, in the initial stage, the daisy chain bidirectional function check is first performed to determine whether the data has PEC anomalies and invalid data. If there are no anomalies, there is no chain break fault; when the collected data has PEC error & invalid CMU data after the node or abnormal end chain break detection, continuous reading is performed; S2, read once in the forward direction and once in the reverse direction, which is counted as one cycle; S3, judging whether the read data is normal and the disconnection information of the erroneous data; S4, execute the above sequence S1 to S3 in a loop, and record the number of loops N; S5, determining whether the number of consecutive cycles N ≥ 3; S6, if N≥3, perform a reset operation, reinitialize, and repeat S1 to S4; S7: If N≥3 is still satisfied, a multi-point fault is reported, and the corresponding broken link position is reported.

4. A highly robust daisy chain break detection method according to claim 3, characterized in that: The S3 judgment process includes: S31, if normal data is read unidirectionally during the loop, the loop is exited and a no-broken-link fault is reported; S32: If the unidirectional data are all wrong and the bidirectional data combination is complete for 10 consecutive cycles, determine whether the error is at the same position. If it is at the same position, determine that the single point is broken and report the broken position, and jump to bidirectional communication. If it is not at the same position, report the single point fault and jump to bidirectional communication. The broken position report is uncertain. If the single point fault disappears during the bidirectional communication, jump to S31. If the bidirectional combination data is incomplete during the bidirectional communication, jump to S33 for judgment. S33, if there are errors in the unidirectional data and the combination of the bidirectional data is incomplete for 10 consecutive cycles, proceed to step S4.

5. A highly robust daisy chain break detection method according to claim 3, characterized in that: The steps of the chain break diagnosis strategy during the system operation phase are as follows: S1, if there is no PEC problem in data transmission and the data is valid, the link break detection will not be triggered; when there is a PEC error in the collected data and the data received by the CMU after the node is invalid, the data validity of the data segment is immediately set to invalid, and the link break detection is started, and continuous data reading begins; S2, read once in the forward direction and once in the reverse direction, which is recorded as one cycle; if the one-way communication data returns to normal during the cycle, the data validity is set to valid; If the data combination is normal during the cycle, the data validity is immediately set to valid; if the forward and reverse data combination is incomplete, the lost data is set to invalid, and the readable data is uploaded normally and the data is valid; S3, judging whether the read data is normal and the disconnection information of the erroneous data; S4, then execute the above sequence S1 to S3 in a loop and record the number of loops N; S5, determining whether the number of consecutive cycles N ≥ 3; S6, if N ≥ 3, perform a reset operation to reinitialize; and repeat S1 to S4; S7: If N≥3 is satisfied for three consecutive cycles, multiple link breaks are determined and the link break locations are reported.

6. A highly robust daisy chain break detection method according to claim 5, characterized in that: The judgment steps in S3 include: S31, if normal data is read unidirectionally during the loop, the loop is exited; S32: If the unidirectional data are all wrong and the bidirectional data combination is complete for 10 consecutive cycles, determine whether the error is at the same position. If it is at the same position, determine that the single point is broken and report the broken position, and jump to bidirectional communication. If it is not at the same position, report the single point fault and jump to bidirectional communication. The broken position report is uncertain. If the single point fault disappears during the bidirectional communication, jump to S31. If the bidirectional combination data is incomplete during the bidirectional communication, jump to S33 for judgment. S33: If there are errors in the unidirectional data and the combination of the bidirectional data is incomplete for 10 consecutive cycles, then proceed to S4.

7. A highly robust daisy chain break detection method according to claim 6, characterized in that: To implement monitoring of the cell status, including clearing all AFE fault interfaces when attempting to recover from a reset.

8. A highly robust daisy chain break detection method according to claim 7, characterized in that: During the process of detecting multiple point broken links and confirming, if the two-way communication data combination is incomplete, the lost data is set to invalid, and the readable data is uploaded normally and the data is valid.

9. A highly robust daisy chain break detection method according to claim 8, characterized in that: When the reset is attempted, all collected data retains the last value and the data validity is set to invalid until the reset is restored, and then the sampling data is updated.

Images