WO2023010790A1 - 一种感测器的测量校正方法、装置及服务器电源 - Google Patents

一种感测器的测量校正方法、装置及服务器电源 Download PDF

Info

Publication number
WO2023010790A1
WO2023010790A1 PCT/CN2021/142858 CN2021142858W WO2023010790A1 WO 2023010790 A1 WO2023010790 A1 WO 2023010790A1 CN 2021142858 W CN2021142858 W CN 2021142858W WO 2023010790 A1 WO2023010790 A1 WO 2023010790A1
Authority
WO
WIPO (PCT)
Prior art keywords
sensor
value
learning mode
resistor
self
Prior art date
Application number
PCT/CN2021/142858
Other languages
English (en)
French (fr)
Inventor
吴名伟
韩红瑞
Original Assignee
苏州浪潮智能科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 苏州浪潮智能科技有限公司 filed Critical 苏州浪潮智能科技有限公司
Priority to US18/259,735 priority Critical patent/US20240061061A1/en
Publication of WO2023010790A1 publication Critical patent/WO2023010790A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/40—Testing power supplies
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R35/00—Testing or calibrating of apparatus covered by the other groups of this subclass
    • G01R35/005—Calibrating; Standards or reference devices, e.g. voltage or resistance standards, "golden" references
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R15/00—Details of measuring arrangements of the types provided for in groups G01R17/00 - G01R29/00, G01R33/00 - G01R33/26 or G01R35/00
    • G01R15/14—Adaptations providing voltage or current isolation, e.g. for high-voltage or high-current networks
    • G01R15/144—Measuring arrangements for voltage not covered by other subgroups of G01R15/14
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
    • G01R19/0084—Arrangements for measuring currents or voltages or for indicating presence or sign thereof measuring voltage only

Definitions

  • the present application relates to the field of server power management, in particular to a sensor measurement and correction method, device and server power supply.
  • the server system needs to read the measured values of various sensors (such as voltage measured value, current measured value, temperature measured value, power measured value, etc.) from the server power supply at any time for power management and system optimization.
  • each sensor in the server power supply transmits its own measurement value to the MCU (Microcontroller Unit, microprocessor) of the server power supply, and the server system uses the BMC (Baseboard Management Controller, baseboard management controller) through the I2C Bus (Inter- Integrated Circuit Bus, serial communication bus) to access the MCU of the server power supply to obtain the readings of various sensors.
  • MCU Microcontroller Unit, microprocessor
  • BMC Baseboard Management Controller
  • I2C Bus Inter- Integrated Circuit Bus, serial communication bus
  • FIG. 1 is a hardware circuit diagram of an internal sensor of a server power supply in the prior art.
  • the sensor includes a shunt resistor and a differential amplifier circuit (differential amplifier + setting resistors R1, R2, R3, R4).
  • the shunt resistor converts the signal corresponding to the detection source into a voltage signal
  • the dynamic amplification circuit amplifies the voltage signal converted by the shunt resistor (the amplification factor is equal to R1/R3), and transmits the amplified voltage signal to the ADC (Analog-to-digital converter, analog-to-digital converter) of the server power MCU , so that the received analog signal is converted into a digital signal by the ADC for internal processing by the MCU.
  • the ADC Analog-to-digital converter, analog-to-digital converter
  • the present application provides a sensor measurement and correction method, which is applied to a sensor including a shunt resistor and a differential amplifier circuit, including:
  • the voltage signal is corrected according to the error value of the shunt resistor and the compensation coefficient to obtain a voltage correction signal for optimal management of the system.
  • the process of pre-compensating and verifying the sensor to obtain the compensation coefficient of the sensor includes:
  • the initial measurement value to be verified of the sensor under different setting values is obtained
  • S cale is the compensation coefficient
  • S cale is the compensation coefficient
  • S cale is the compensation coefficient
  • the process of correcting the voltage signal according to the error value of the shunt resistor and the compensation coefficient to obtain the voltage correction signal includes:
  • Vo_real is the voltage correction signal
  • ADC Count is the digital signal value corresponding to the voltage signal
  • Rshunt Count is the digital signal value corresponding to the error value of the shunt resistor
  • V Gain is the differential amplifier circuit gain value.
  • the measurement and correction method of the sensor further includes:
  • Update the relationship according to the preset compensation coefficient Calculating the updated compensation coefficient of the sensor to correct the voltage signal according to the updated compensation coefficient; wherein, S' cale is the updated compensation coefficient; is the digital signal value corresponding to the new measured value to be verified; It is the actual digital signal value of the parameter to be measured under the current load setting value.
  • the process of triggering the automatic self-learning mode includes:
  • the measured value to be verified of the sensor under the current load setting value is obtained multiple times, and the average value to be verified of the obtained multiple measured values to be verified is calculated according to the moving average algorithm measured value;
  • the automatic self-learning mode is triggered.
  • the process of triggering the automatic self-learning mode includes:
  • the automatic self-learning mode After receiving the instruction to enter the automatic self-learning mode, the automatic self-learning mode is triggered; wherein, the condition for issuing the instruction to enter the automatic self-learning mode is: the sensor to be calibrated under the current load setting value obtained multiple times The difference between the average value of the test measurement value and the median value of the measurement values to be verified obtained multiple times does not reach the preset bias threshold.
  • the measurement and correction method of the sensor further includes:
  • the self-learning mode includes a manual self-learning mode and an automatic self-learning mode
  • the measurement and correction method of the sensor further includes:
  • the measured values to be verified obtained in manual self-learning mode and automatic self-learning mode are all stored in the memory in the device, so as to determine the total number of self-learning by querying the storage content of the memory, which is used to manage the capacity of memory resources.
  • the embodiment of the present application also provides a measurement and calibration device for a sensor, including a memory and one or more processors, where computer-readable instructions are stored in the memory, and when the computer-readable instructions are executed by one or more processors, One or more processors are made to execute the steps of any one of the sensor measurement correction methods described above.
  • the embodiment of the present application also provides one or more non-volatile computer-readable storage media storing computer-readable instructions.
  • the computer-readable instructions are executed by one or more processors, the one or more processors Execute the steps of any one of the above-mentioned sensor measurement calibration methods.
  • the present application also provides a server power supply, including a sensor and a measurement and correction device for the sensor; the sensor includes a shunt resistor and includes a differential amplifier, a first resistor, a second resistor, and a third resistor And the differential amplifier circuit of the fourth resistor; wherein:
  • the input negative terminal of the differential amplifier is respectively connected to the first terminal of the first resistor and the first terminal of the third resistor, and the second terminal of the first resistor is connected to the output terminal of the differential amplifier.
  • the second end of the third resistor is connected to the current output end of the shunt resistor
  • the positive input end of the differential amplifier is respectively connected to the first end of the second resistor and the fourth resistor
  • the first end is connected, the second end of the second resistor is connected to the reference end of the differential amplifier, and the second end of the fourth resistor is connected to the current input end of the shunt resistor;
  • the ratio of the third resistor to the first resistor is equal to the ratio of the fourth resistor to the second resistor; the ratio of the first resistor to the third resistor is adjusted to adjust the The gain value of the differential amplifier circuit.
  • Fig. 1 is a hardware circuit diagram of a sensor inside a server power supply in the prior art
  • FIG. 2 is a flow chart of a sensor measurement and correction method provided by the present application according to one or more embodiments
  • FIG. 3 is a schematic diagram of a current parameter provided by the present application according to one or more embodiments.
  • Fig. 4 is a schematic diagram of compensation and adjustment of a current parameter provided by the present application according to one or more embodiments;
  • Fig. 5 is a comparison graph of a sensor reading value before and after correction provided by the present application according to one or more embodiments
  • FIG. 6 is a structure diagram of a Wheatbridge bridge provided by the present application according to one or more embodiments.
  • FIG. 7 is an equivalent circuit diagram of a sensor architecture provided by the present application according to one or more embodiments.
  • Fig. 8 is a schematic structural diagram of a sensor measurement and calibration device according to one or more embodiments of the present application.
  • the core of this application is to provide a sensor measurement and correction method, device and server power supply, which can improve the reading accuracy by correcting the voltage signal output by the sensor through software, that is, there is no need to select high-precision shunt resistors and low-voltage
  • a shifted differential amplifier also enables improved read accuracy, saving the overall cost of the sensor in the server power supply.
  • FIG. 2 is a flow chart of a sensor measurement and calibration method provided by an embodiment of the present application.
  • the measurement correction method for this sensor is applied to a sensor containing a shunt resistor and a differential amplifier circuit, including:
  • Step S1 Perform compensation calibration on the sensor in advance to obtain the compensation coefficient of the sensor.
  • Step S2 Obtain the voltage signal output by the sensor.
  • Step S3 Correcting the voltage signal according to the error value of the shunt resistor and the compensation coefficient to obtain a voltage correction signal for use in system optimization management.
  • the present application performs compensation verification on the sensor in advance to obtain the compensation coefficient of the sensor, which is used for subsequent correction of the voltage signal output by the sensor.
  • This application obtains the voltage signal output by the sensor, and after obtaining the voltage signal output by the sensor, corrects the voltage signal output by the sensor according to the error value of the shunt resistor and the compensation coefficient of the sensor to obtain the voltage Correction signals for system optimization management.
  • the application can improve the reading accuracy by correcting the voltage signal output by the sensor through software, that is, the reading accuracy can be improved without selecting a high-precision shunt resistor and a differential amplifier with low voltage drift, thereby saving The overall cost of sensors within the server power supply.
  • the process of pre-compensating and verifying the sensor to obtain the compensation coefficient of the sensor includes:
  • the initial measurement value to be verified of the sensor under different setting values is obtained
  • S cale is the compensation coefficient
  • S cale is the compensation coefficient
  • S cale is the compensation coefficient
  • this application can use the manual self-learning mode to learn the compensation coefficient of the sensor for the compensation verification of the sensor in advance.
  • the system can issue an instruction to enter the manual self-learning mode. Enter the manual self-learning mode after the instruction to enter the manual self-learning mode:
  • This application can be used with an electronic load machine to manually adjust the load of the device where the sensor is located (such as a server power supply).
  • the electronic load machine is set to 0% load, 50% load or 100% load.
  • the present application obtains the measurement value of the sensor under different setting values (referred to as the initial measurement value to be verified).
  • the preset compensation coefficient relational formula can be used Compute the compensation coefficients for the sensor, where, Indicates the actual value of the parameter to be measured corresponding to the sensor under the first load setting value; Indicates the actual value of the parameter to be measured corresponding to the sensor under the second load setting value; Indicates the initial measurement value to be verified of the sensor under the first load setting value; Indicates the initial measurement value to be verified of the sensor under the second load setting value; Both are digital signal values converted by ADC.
  • the process of correcting the voltage signal according to the error value of the shunt resistor and the compensation coefficient to obtain the voltage correction signal includes:
  • Vo_real is the voltage correction signal
  • ADC Count is the digital signal value corresponding to the voltage signal
  • Rshunt Count is the digital signal value corresponding to the error value of the shunt resistor
  • V Gain is the differential amplifier circuit gain value.
  • the correction relational expression of the voltage signal output by the sensor is: Among them, ADC Count represents the digital signal value converted by the ADC from the voltage signal output by the sensor; Rshunt Count represents the digital signal value converted by the ADC from the error value of the shunt resistor, because the error value of the shunt resistor will cause the sensor output
  • the error of the voltage signal so this parameter can be used to deduct the error, generally the error that needs to be compensated is 1%-3%;
  • V Gain is the gain value of the differential amplifier circuit, which is equal to R1/R3; S cale is the compensation of the sensor coefficient.
  • the current range is 0A-25A, and a parameter compensation adjustment is made.
  • the current parameters are shown in Figure 3, then Compensate and adjust the current parameters of 0A-25A to obtain the results shown in Figure 4, and obtain the comparison curve of the sensor reading value before and after correction as shown in Figure 5 (the horizontal axis represents the actual current value/A ;
  • the vertical axis represents the current reading value/A) of the sensor, wherein the reference curve is an ideal reading value curve without any error, the curve before calibration is the sensor reading value without firmware calibration, and the curve after calibration is the sensor reading value after calibration. Sensor readings corrected by firmware. It can be seen from this example that the reading error can be effectively compensated after firmware calibration and is close to the ideal value. After 10% load (3A), the accuracy of the reading segment can reach within 1%.
  • the measurement and correction method of the sensor also includes:
  • Update the relationship according to the preset compensation coefficient Calculating the updated compensation coefficient of the sensor to correct the voltage signal according to the updated compensation coefficient; wherein, S' cale is the updated compensation coefficient; is the digital signal value corresponding to the new measured value to be verified; It is the actual digital signal value of the parameter to be measured under the current load setting value.
  • the reading value deviation is caused by the aging of the device components of the sensor, and the sensor calculated based on the initial measured value of the sensor to be verified
  • the compensation coefficient of the sensor is not accurate enough.
  • the automatic self-learning mode can be triggered to automatically learn the accurate compensation coefficient of the sensor.
  • the process of triggering the automatic self-learning mode includes:
  • the measured value to be verified of the sensor under the current load setting value is obtained multiple times, and the average value to be verified of the obtained multiple measured values to be verified is calculated according to the moving average algorithm Measurements;
  • the automatic self-learning mode is triggered.
  • the step of triggering the automatic self-learning mode is executed based on the determination result .
  • the first way to trigger the automatic self-learning mode is (self-triggering): when the device where the sensor is located is running, collect the pending calibration of the sensor for multiple times (such as 16 times) under the current load setting value.
  • SMA moving average
  • the average measured value to be verified of the multiple measured values to be verified is obtained, and then the average measured value to be verified is compared with the sensor under the current load setting value
  • the difference between the corresponding initial measurement values to be verified is determined to determine whether the difference between the two is greater than the preset error threshold. If the difference between the two is not greater than the preset error threshold, it means that the current compensation coefficient of the sensor is accurate enough.
  • Trigger the automatic self-learning mode if the difference between the two is greater than the preset error threshold, indicating that the current compensation coefficient of the sensor is not accurate enough, the automatic self-learning mode will be triggered. It should be noted that the SMA algorithm can ensure that the change of the average value is consistent with the change of the data, rather than changing with time, and it is not easy to cause false actions due to noise.
  • electric meters are divided into seven grades: 0.1, 0.2, 0.5, 1.0, 1.5, 2.5, and 5.0.
  • grade value the higher the accuracy of the meter.
  • Commonly used meters have three accuracy levels: 0.5S, 1st, and 2nd.
  • Level 2 means that the allowable error of the meter is within ⁇ 2%, and the smaller the value of the level, the higher the accuracy of the meter.
  • the single-phase meters installed by the State Grid for users are all grade 2, and the allowable error is within ⁇ 2%, so the preset error threshold here can be set to 2%, or the preset error threshold can be based on other levels (such as 1.0, 1.5 , 2.5, 5.0 four levels) setting.
  • the process of triggering the automatic self-learning mode includes:
  • the automatic self-learning mode After receiving the instruction to enter the automatic self-learning mode, the automatic self-learning mode is triggered; wherein, the condition for issuing the instruction to enter the automatic self-learning mode is: the sensor to be calibrated under the current load setting value obtained multiple times The difference between the average value of the test measurement value and the median value of the measurement values to be verified obtained multiple times does not reach the preset bias threshold.
  • the second way to trigger the automatic self-learning mode is (command trigger): the system (such as server BMC) will read the sensor readings in the device (such as server power supply) at any time to optimize system management.
  • the system can use the mean value method of the probability density function of normal distribution to judge the standard deviation, that is, obtain the measurement value to be verified of the sensor under the current load setting value multiple times (such as 16 times), and obtain the multiple acquisition
  • the average value of the measured values to be verified and calculate the difference between the average value of the measured values to be verified obtained multiple times and the median value of the measured values to be verified obtained multiple times, if the difference reaches the preset
  • the error threshold indicates that the current compensation coefficient of the sensor is accurate enough, and the system will not issue an instruction to enter the automatic self-learning mode; if the difference does not reach the preset error threshold, it indicates that the current compensation coefficient of the sensor is not accurate enough. Then the system issues an instruction to enter the automatic self-learning mode.
  • the application triggers the automatic self-learning mode after receiving the
  • the standard deviation of 1 standard deviation (1 ⁇ ) is 31.73%
  • 2 standard deviations (2 ⁇ ) is 4.55%
  • 2.5 standard deviations (2.5 ⁇ ) is 1%
  • the standard deviation (3 ⁇ ) bias is 0.2%. Since the average value of the measured values to be verified obtained multiple times needs to be 2.5 standard deviations higher than the median value of the measured values to be verified obtained multiple times, the preset error threshold here is selected as 1%.
  • the measurement and correction method of the sensor also includes:
  • the self-learning mode includes a manual self-learning mode and an automatic self-learning mode
  • D1h Define D1h as entering the self-learning mode: if the PSU PMBus command address D1h Bit 0 is written to 1, it will start to enter the automatic self-learning mode. If the PSU PMBus command address D1h Bit 0 is written to 0, it will be forced to jump out of the automatic self-learning mode. If the PSU PMBus command address D1h Bit 1 is written to 1, it will start to enter the manual self-learning mode. If the PSU PMBus command address D1h Bit 1 is written to 0, it will force to jump out of the manual self-learning mode.
  • the industrial computer sends a command address D1h Bit 1 to the PSU and writes 1, and the PSU starts to enter the manual self-learning mode.
  • the industrial computer waits for the PSU to return whether the D2h Bit 1 is 1. If it is 1, it is determined that the PSU has officially entered the manual self-learning mode.
  • the electronic load machine is set to 0% load, and then wait for 4 seconds, and the PSU records the measurement value to be verified when the sensor is at 0% load.
  • the LED (light-emitting diode, light-emitting diode) of the server PSU keeps flashing green.
  • the electronic load machine is set to 100% load, and then wait for 4 seconds, and the PSU records the measurement value to be verified when the sensor is at 100% load. At this time, the LED of the server PSU keeps blinking green.
  • the process of the server PSU self-starting automatic self-learning mode is as follows:
  • the server PSU compares the sensor reading value under the current load setting value with the initial setting reading value, if the error between the two is greater than ⁇ 2%, it will start to execute the automatic self-learning mode.
  • the server PSU will send a D2h position equal to 1 notification system to inform the system that the PSU has officially entered the automatic self-learning mode.
  • the server PSU will use the current system load as the calibration value, and then wait for 4 seconds. After the server PSU calculates the digital quantity of the measured value to be calibrated, it will store the digital quantity of the measured value to be calibrated to the MCU of the PSU EEPROM, but will not overwrite the original factory settings, so as to learn the accurate compensation coefficient of the sensor. At this time, the LED of the server PSU is continuously lit green.
  • the process for the server BMC to start the automatic self-learning mode of the server PSU is as follows:
  • the server BMC compares the sensor reading value under the current load setting value to be lower than 2.5 standard deviations, it will control the server PSU to start the automatic self-learning mode.
  • the server PSU sends a D2h position equal to 1 notification system through the IPMI (Intelligent Platform Management Interface) command of the BMC to inform the system that the PSU has officially entered the automatic self-learning mode.
  • IPMI Intelligent Platform Management Interface
  • the server PSU will use the current system load as the calibration value, and then wait for 4 seconds. After the server PSU calculates the digital quantity of the measured value to be calibrated, it will store the digital quantity of the measured value to be calibrated to the MCU of the PSU EEPROM, but will not overwrite the original factory settings, so as to learn the accurate compensation coefficient of the sensor. At this time, the LED of the server PSU is continuously lit green.
  • the measurement and correction method of the sensor also includes:
  • the measured values to be verified obtained in the manual self-learning mode and the automatic self-learning mode are all stored in the memory in the device, so that the total number of self-learning can be determined by querying the storage content of the memory, which is used to manage the capacity resources of the memory.
  • the above-mentioned embodiments have mentioned that the measured values to be verified obtained in the manual self-learning mode and the automatic self-learning mode can be stored in the memory in the device.
  • the system can query the memory in the device
  • the storage content can determine the total number of self-learning times (manual self-learning times + automatic self-learning times) for use by the capacity resources of the system management memory.
  • the measurement and correction methods for the above sensor can be implemented by the MCU in the server PSU.
  • the server PSU will use the MCU to complete the functions such as converter switch control, fan control, LED control, monitoring, protection and communication in the power supply.
  • the MCU will be divided into primary side MCU and secondary side MCU. Since the secondary test MCU is mainly used for communication with the BMC, the measured value to be verified will be stored in the EEPROM of the secondary test MCU.
  • the server BMC can query the registers of the MCU EEPROM in the server PSU through the IPMI command to know whether the server PSU has performed self-learning mode and know the total number of self-learning.
  • this application uses the Wheatbridge balance method to derive and simplify the model of the differential amplifier:
  • the Wheatbridge bridge includes the resistance to be measured Rx and the variable resistor R2 of the known resistance, the resistance R1 and the resistance R3, connect R1 and R2 in series, R3 and Rx in series, and then connect the two series
  • the circuits are connected in parallel, a wire is connected between the midpoint of the wire between R1 and R2 and the midpoint of the wire between R3 and Rx, and a galvanometer V G is placed on this wire.
  • the setting resistors R1, R2, R3, and R4 are usually several K ohm (kilohm), so the shunt resistor can be equivalent to a short circuit, so the setting The fixed resistors R3 and R4 can be connected.
  • the REF reference voltage and the output terminal OUT of the differential amplifier can be equivalent to a short circuit, so the setting resistors R1 and R2 can be connected.
  • the signal amplification factor of the differential amplifier can be improved by adjusting R1 and R3.
  • the present application also provides a sensor measurement calibration device, the sensor measurement calibration device may include computer equipment, the computer equipment may be a terminal or a server, the sensor measurement calibration device
  • the internal structure diagram of the device can be shown in FIG. 8 .
  • the measuring and correcting device for a sensor includes a processor, a memory, a network interface and an input device connected through a system bus. Among them, the processor is used to provide calculation and control capabilities.
  • the memory includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium stores an operating system and computer readable instructions.
  • the internal memory provides an environment for the execution of the operating system and computer readable instructions in the non-volatile storage medium.
  • the network interface of the computer device is used to communicate with an external terminal or server through a network connection.
  • the input device may be a touch layer covered on the display screen, or a button, a trackball or a touch pad provided on the casing of the computer equipment, or an external keyboard, touch pad or mouse.
  • FIG. 8 is only a block diagram of a part of the structure related to the solution of this application, and does not constitute a limitation on the equipment to which the solution of this application is applied.
  • the specific equipment may include More or fewer components are shown in the figures, or certain components are combined, or have different component arrangements.
  • the embodiment of the present application also provides a non-volatile readable storage medium, where computer-readable instructions are stored in the non-volatile readable storage medium, and when the computer-readable instructions are executed by one or more processors, the The steps of the method for measuring and correcting the sensor in any one of the above embodiments are realized.
  • Nonvolatile memory can include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory can include random access memory (RAM) or external cache memory.
  • RAM random access memory
  • RAM is available in many forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Chain Synchlink DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
  • the present application also provides a server power supply, including a sensor and a measurement and correction device for the sensor; the sensor includes a shunt resistor and includes a differential amplifier, a first resistor, a second resistor, a third resistor, and a first resistor.
  • a server power supply including a sensor and a measurement and correction device for the sensor; the sensor includes a shunt resistor and includes a differential amplifier, a first resistor, a second resistor, a third resistor, and a first resistor.
  • the input negative terminal of the differential amplifier is respectively connected to the first terminal of the first resistor and the first terminal of the third resistor, the second terminal of the first resistor is connected to the output terminal of the differential amplifier, and the second terminal of the third resistor is connected to the output terminal of the differential amplifier.
  • the current output terminal of the shunt resistor is connected, the input positive terminal of the differential amplifier is respectively connected with the first terminal of the second resistor and the first terminal of the fourth resistor, and the second terminal of the second resistor is connected with the reference terminal of the differential amplifier , the second end of the fourth resistor is connected to the current input end of the shunt resistor;
  • the ratio of the third resistor to the first resistor is equal to the ratio of the fourth resistor to the second resistor; the gain value of the differential amplifier circuit is adjusted by adjusting the ratio of the first resistor to the third resistor.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Arrangements For Transmission Of Measured Signals (AREA)
  • Measurement Of Current Or Voltage (AREA)
  • Testing Or Calibration Of Command Recording Devices (AREA)

Abstract

一种感测器的测量校正方法、装置及服务器电源，应用于包含分流电阻器和差动放大器的感测器，包括：预先对感测器进行补偿校验，得到感测器的补偿系数（步骤S1）；获取感测器输出的电压信号（步骤S2）；根据分流电阻器的误差值和补偿系数校正电压信号，得到电压校正信号，以供系统优化管理使用（步骤S3）。

Description

一种感测器的测量校正方法、装置及服务器电源
相关申请的交叉引用
本申请要求于2021年8月5日提交中国专利局，申请号为202110893892.2，申请名称为“一种感测器的测量校正方法、装置及服务器电源”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及服务器电源管理领域，特别是涉及一种感测器的测量校正方法、装置及服务器电源。
背景技术
服务器系统需要随时从服务器电源中读取其内各感测器的测量值(如电压测量值、电流测量值、温度测量值、功率测量值等)，以便进行电源管理与系统优化。目前，服务器电源内各感测器将自身的测量值传输至服务器电源的MCU(Microcontroller Unit，微处理器)，服务器系统利用BMC(Baseboard Management Controller，基板管理控制器)透过I2C Bus(Inter-Integrated Circuit Bus，串行通讯总线)访问服务器电源的MCU，以获取各项感测器读值。
请参照图1，图1为现有技术中的一种服务器电源内部感测器的硬件线路图。感测器包括分流电阻器及差动放大电路(差动放大器+设定电阻R1、R2、R3、R4)，其工作原理为：分流电阻器将对应侦测源的信号转换为电压信号，差动放大电路将分流电阻器转换的电压信号进行放大处理(放大倍数等于R1/R3)，并将放大后的电压信号传输至服务器电源MCU的ADC(Analog-to-digital converter，模数转换器)，以由ADC将接收的模拟信号转换为数字信号后供MCU内部处理。一般情况下，服务器电源内大约需5-10个感测器来完成整个电源管理所需信号的侦测。
但是，感测器内分流电阻器及设定电阻均会存在电阻误差，差动放大器会存在电压飘移，导致服务器电源的MCU从感测器中读取的电压信号存在误差，使得读取精度较低。目前，为了提高读取精度，通常挑选高精度的分流电阻器、设定电阻及低电压飘移的差动放大器，但是，高精度的分流电阻器及低电压飘移的差动放大器非常昂贵，导致服务器电源内感测器的整体成本较高。
因此，如何提供一种解决上述技术问题的方案是本领域的技术人员目前需要解决的问题。
发明内容
本申请提供了一种感测器的测量校正方法，应用于包含分流电阻器和差动放大电路的感测器，包括：
预先对感测器进行补偿校验，得到感测器的补偿系数；
获取感测器输出的电压信号；及
根据分流电阻器的误差值和补偿系数校正电压信号，得到电压校正信号，以供系统优化管理使用。
在其中一个实施例中，预先对感测器进行补偿校验，得到感测器的补偿系数的过程，包括：
在接收到进入手动自我学习模式的指令后，进入手动自我学习模式；
根据感测器所在设备的负载的不同设定值，得到感测器在不同设定值下的初始待校验测量值；
根据预设补偿系数关系式
Figure PCTCN2021142858-appb-000001
计算感测器的补偿系数；其中，S cale为补偿系数；
Figure PCTCN2021142858-appb-000002
为感测器对应的待测参数在第一负载设定值下的实际数字信号值；
Figure PCTCN2021142858-appb-000003
为待测参数在第二负载设定值下的实际数字信号值；
Figure PCTCN2021142858-appb-000004
为感测器在第一负载设定值下的初始待校验测量值对应的数字信号值；
Figure PCTCN2021142858-appb-000005
为感测器在第二负载设定值下的初始待校验测量值对应的数字信号值。
在其中一个实施例中，根据分流电阻器的误差值和补偿系数校正电压信号，得到电压校正信号的过程，包括：
根据预设电压校正关系式
Figure PCTCN2021142858-appb-000006
校正电压信号，得到电压校正信号；其中，Vo_real为电压校正信号；ADC Count为电压信号对应的数字信号值；Rshunt Count为分流电阻器的误差值对应的数字信号值；V Gain为差动放大电路的增益值。
在其中一个实施例中，感测器的测量校正方法还包括：
在触发自动自我学习模式时，进入自动自我学习模式；
获取感测器在当前负载设定值下的新待校验测量值；及
根据预设补偿系数更新关系式
Figure PCTCN2021142858-appb-000007
计算感测器更新的补偿系数，以根据更新的补偿系数校正电压信号；其中，S′ cale为更新的补偿系数；
Figure PCTCN2021142858-appb-000008
为新待校验测量值对应的数字信号值；
Figure PCTCN2021142858-appb-000009
为待测参数在当前负载设定值下的实际数字信号值。
在其中一个实施例中，触发自动自我学习模式的过程，包括：
在感测器所在设备运行时，多次获取感测器在当前负载设定值下的待校验测量值，并根据移动平均算法求取获取的多个待校验测量值的平均待校验测量值；及
在平均待校验测量值与感测器在当前负载设定值下对应的初始待校验测量值的差值大于预设误差阈值时，触发自动自我学习模式。
在其中一个实施例中，触发自动自我学习模式的过程，包括：
在接收到进入自动自我学习模式的指令后，触发自动自我学习模式；其中，进入自动自我学习模式的指令的下发条件为：多次获取的感测器在当前负载设定值下的待校验测量值的平均值高于多次获取的待校验测量值的中间值的差值未达到预设偏误阈值。
在其中一个实施例中，感测器的测量校正方法还包括：
预先为设备内待定义功能的第一指令地址和第二指令地址分别定义进入自我学习模式的功能和自我学习模式状态回报的功能；其中，自我学习模式包括手动自我学习模式和自动自我学习模式；
在接收到进入手动自我学习模式或自动自我学习模式的指令时，相应向第一指令地址写入表示进入手动自我学习模式的设定值或表示进入自动自我学习模式的设定值，以开始进入手动自我学习模式或自动自我学习模式；及
根据当前自我学习模式的学习状态，向第二指令地址写入相应的设定值，并将第二指令地址写入相应的设定值的信息反馈至系统。
在其中一个实施例中，感测器的测量校正方法还包括：
将在手动自我学习模式和自动自我学习模式下获取的待校验测量值均存储至设备内的存储器中，以通过查询存储器的存储内容确定自我学习总次数，供管理存储器的容量 资源使用。
本申请实施例还提供了一种感测器的测量校正装置，包括存储器及一个或多个处理器，存储器中储存有计算机可读指令，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行上述任一项感测器的测量校正方法的步骤。
本申请实施例最后还提供了一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行上述任一项感测器的测量校正方法的步骤。
本申请还提供了一种服务器电源，包括感测器及上述感测器的测量校正装置；所述感测器包括分流电阻器及包含差动放大器、第一电阻、第二电阻、第三电阻及第四电阻的差动放大电路；其中：
所述差动放大器的输入负端分别与所述第一电阻的第一端和所述第三电阻的第一端连接，所述第一电阻的第二端与所述差动放大器的输出端连接，所述第三电阻的第二端与所述分流电阻器的电流输出端连接，所述差动放大器的输入正端分别与所述第二电阻的第一端和所述第四电阻的第一端连接，所述第二电阻的第二端与所述差动放大器的参考端连接，所述第四电阻的第二端与所述分流电阻器的电流输入端连接；
其中，所述第三电阻与所述第一电阻的比值等于所述第四电阻与所述第二电阻的比值；通过调整所述第一电阻和所述第三电阻的比值大小来调整所述差动放大电路的增益值。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对现有技术和实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为现有技术中的一种服务器电源内部感测器的硬件线路图；
图2为本申请根据一个或多个实施例提供的一种感测器的测量校正方法的流程图；
图3为本申请根据一个或多个实施例提供的一种电流参数的示意图；
图4为本申请根据一个或多个实施例提供的一种电流参数的补偿调整示意图；
图5为本申请根据一个或多个实施例提供的一种感测器读值校正前和校正后的对比 曲线图；
图6为本申请根据一个或多个实施例提供的一种惠式电桥架构图；
图7为本申请根据一个或多个实施例提供的一种感测器架构等效电路图；
图8为本申请根据一个或多个实施例提供的一种感测器的测量校正装置的结构示意图。
具体实施方式
本申请的核心是提供一种感测器的测量校正方法、装置及服务器电源，可通过软件校正感测器输出的电压信号来提高读取精度，即无需选择高精度的分流电阻器及低电压飘移的差动放大器也能实现读取精度的提高，从而节约了服务器电源内感测器的整体成本。
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
请参照图2，图2为本申请实施例提供的一种感测器的测量校正方法的流程图。
该感测器的测量校正方法应用于包含分流电阻器和差动放大电路的感测器，包括：
步骤S1：预先对感测器进行补偿校验，得到感测器的补偿系数。
步骤S2：获取感测器输出的电压信号。
步骤S3：根据分流电阻器的误差值和补偿系数校正电压信号，得到电压校正信号，以供系统优化管理使用。
具体地，本申请提前对感测器进行补偿校验，得到感测器的补偿系数，以为后续校正感测器输出的电压信号使用。本申请获取感测器输出的电压信号，并在获取到感测器输出的电压信号之后，根据分流电阻器的误差值和感测器的补偿系数，校正感测器输出的电压信号，得到电压校正信号，以供系统优化管理使用。
可见，本申请可通过软件校正感测器输出的电压信号来提高读取精度，即无需选择高精度的分流电阻器及低电压飘移的差动放大器也能实现读取精度的提高，从而节约了服务器电源内感测器的整体成本。
在上述实施例的基础上：
作为一种可选的实施例，预先对感测器进行补偿校验，得到感测器的补偿系数的过程，包括：
在接收到进入手动自我学习模式的指令后，进入手动自我学习模式；
根据感测器所在设备的负载的不同设定值，得到感测器在不同设定值下的初始待校验测量值；
根据预设补偿系数关系式
Figure PCTCN2021142858-appb-000010
计算感测器的补偿系数；其中，S cale为补偿系数；
Figure PCTCN2021142858-appb-000011
为感测器对应的待测参数在第一负载设定值下的实际数字信号值；
Figure PCTCN2021142858-appb-000012
为待测参数在第二负载设定值下的实际数字信号值；
Figure PCTCN2021142858-appb-000013
为感测器在第一负载设定值下的初始待校验测量值对应的数字信号值；
Figure PCTCN2021142858-appb-000014
为感测器在第二负载设定值下的初始待校验测量值对应的数字信号值。
具体地，本申请提前对感测器进行的补偿校验可以使用手动自我学习模式，以学习到感测器的补偿系数，此时系统可下发进入手动自我学习模式的指令，本申请在接收到进入手动自我学习模式的指令后，进入手动自我学习模式：
本申请可搭配电子负载机手动调节感测器所在设备(如服务器电源)的负载，如电子负载机设定为0％负载或50％负载或100％负载。本申请根据感测器所在设备的负载的不同设定值，得到感测器在不同设定值下的测量值(称为初始待校验测量值)。基于感测器在不同设定值下的初始待校验测量值，可根据预设补偿系数关系式
Figure PCTCN2021142858-appb-000015
计算感测器的补偿系数，其中，
Figure PCTCN2021142858-appb-000016
表示感测器对应的待测参数在第一负载设定值下的实际值；
Figure PCTCN2021142858-appb-000017
表示感测器对应的待测参数在第二负载设定值下的实际值；
Figure PCTCN2021142858-appb-000018
表示感测器在第一负载设定值下的初始待校验测量值；
Figure PCTCN2021142858-appb-000019
表示感测器在第二负载设定值下的初始待校验测量值；
Figure PCTCN2021142858-appb-000020
Figure PCTCN2021142858-appb-000021
均为经ADC转换的数字信号值。
作为一种可选的实施例，根据分流电阻器的误差值和补偿系数校正电压信号，得到电压校正信号的过程，包括：
根据预设电压校正关系式
Figure PCTCN2021142858-appb-000022
校正电压信号，得到电压校正信号；其中，Vo_real为电压校正信号；ADC Count为电压信号对应的数字信号值；Rshunt Count为分流电阻器的误差值对应的数字信号值；V Gain为差动放大电路的增益值。
具体地，感测器输出的电压信号的校正关系式为：
Figure PCTCN2021142858-appb-000023
其中，ADC Count表示感测器输出的电压信号经ADC转换的数字信号值；Rshunt Count表示分流电阻器的误差值经ADC转换的数字信号值，因为分流电阻器的误差值会造成感测器输出的电压信号的误差，因此可以利用此参数扣除误差，一般需要补偿的误差为1％-3％；V Gain为差动放大电路的增益值，等于R1/R3；S cale为感测器的补偿系数。需要说明的是，ADC转换的数字信号值最大值为2 n，也是ADC所在的微处理器可提供的最大分辨率，如果为n＝10，即2 10＝1024，最多可切为1023次(0不算)。
更具体地，以一个电流参数校正为例，电流范围为0A-25A，做一个参数补偿调整。电流参数如图3所示，则
Figure PCTCN2021142858-appb-000024
对0A-25A的电流参数进行补偿调整，得到如图4所示的结果，且得到如图5所示的感测器读值校正前和校正后的对比曲线(横轴表示实际电流值/A；纵轴表示感测器的电流读值/A)，其中，基准曲线为理想不受任何误差的读值曲线，校正前曲线为不经固件校正的感测器读值，校正后曲线为经固件校正的感测器读值。可以透过此例得知，经固件校正后可有效补偿读取误差且接近理想值，在10％负载(3A)后，读取段精度可达1％以内。
作为一种可选的实施例，感测器的测量校正方法还包括：
在触发自动自我学习模式时，进入自动自我学习模式；
获取感测器在当前负载设定值下的新待校验测量值；
根据预设补偿系数更新关系式
Figure PCTCN2021142858-appb-000025
计算感测器更新的补偿系数，以根据更新的补偿系数校正电压信号；其中，S′ cale为更新的补偿系数；
Figure PCTCN2021142858-appb-000026
为新待校验测量值对应的数字信号值；
Figure PCTCN2021142858-appb-000027
为待测参数在当前负载设定值下的实际数字信号值。
进一步地，在感测器所在设备运转一段时间后(例如3-5年)，因感测器的器件组件老化造成读值偏差，基于感测器的初始待校验测量值计算的感测器的补偿系数不够准确，此时可触发自动自我学习模式，以自动学习到感测器准确的补偿系数。
具体地，在触发自动自我学习模式时，进入自动自我学习模式：
基于感测器所在设备当前的负载设定值(一般不会发生变化)，获取感测器在当前负载设定值下的新待校验测量值，然后根据预设补偿系数更新关系式
Figure PCTCN2021142858-appb-000028
计算感测器更新的补偿系数，以根据更新的补偿系数校正感测器输出的电压信号
Figure PCTCN2021142858-appb-000029
从而补 偿感测器的器件组件老化造成的读值偏差。其中，
Figure PCTCN2021142858-appb-000030
表示感测器对应的待测参数在当前负载设定值下的实际值；
Figure PCTCN2021142858-appb-000031
表示感测器在当前负载设定值下的新待校验测量值；
Figure PCTCN2021142858-appb-000032
均为经ADC转换的数字信号值。
作为一种可选的实施例，触发自动自我学习模式的过程，包括：
在感测器所在设备运行时，多次获取感测器在当前负载设定值下的待校验测量值，并根据移动平均算法求取获取的多个待校验测量值的平均待校验测量值；
判断平均待校验测量值与感测器在当前负载设定值下对应的初始待校验测量值的差值是否大于预设误差阈值；
若是，则触发自动自我学习模式。
具体地，判定平均待校验测量值与感测器在当前负载设定值下对应的初始待校验测量值的差值大于预设误差阈值，基于该判定结果执行触发自动自我学习模式的步骤。
具体地，第一种触发自动自我学习模式的方式为(自主触发)：在感测器所在设备运行时，连续多次(如16次)采集感测器在当前负载设定值下的待校验测量值，并根据SMA(移动平均)算法求取获取的多个待校验测量值的平均待校验测量值，然后将平均待校验测量值与感测器在当前负载设定值下对应的初始待校验测量值作差，判断二者的差值是否大于预设误差阈值，若二者的差值不大于预设误差阈值，说明感测器当前的补偿系数够准确，则不触发自动自我学习模式；若二者的差值大于预设误差阈值，说明感测器当前的补偿系数不够准确，则触发自动自我学习模式。需要说明的是，SMA算法可以确保平均值的变化与数据的变化一致，而不是随时间变化，且不易因杂讯造成误动作。
目前，规定电表分为七个等级：0.1、0.2、0.5、1.0、1.5、2.5、5.0级。等级数值越小，电表的精确度越高。常用电表有0.5S、1级、2级三个精确度等级。2级则表示电表允许误差在±2％以内，等级数值越小，电表的精确度越高。国网给用户安装的单相电表都是2级，允许误差在±2％以内，所以这里的预设误差阈值可设为2％，也可将预设误差阈值基于其它等级(如1.0、1.5、2.5、5.0四个等级)设定。
作为一种可选的实施例，触发自动自我学习模式的过程，包括：
在接收到进入自动自我学习模式的指令后，触发自动自我学习模式；其中，进入自动自我学习模式的指令的下发条件为：多次获取的感测器在当前负载设定值下的待校验测量值的平均值高于多次获取的待校验测量值的中间值的差值未达到预设偏误阈值。
具体地，第二种触发自动自我学习模式的方式为(指令触发)：系统(如服务器BMC)会随时读取设备(如服务器电源)内感测器读值，来做系统管理优化。此时系统 可采用常态分布的机率密度函数均值法来判断标准偏误，即多次(如16次)获取感测器在当前负载设定值下的待校验测量值，求取多次获取的待校验测量值的平均值，并求取多次获取的待校验测量值的平均值高于多次获取的待校验测量值的中间值的差值，若此差值达到预设偏误阈值，说明感测器当前的补偿系数够准确，则系统不下发进入自动自我学习模式的指令；若此差值未达到预设偏误阈值，说明感测器当前的补偿系数不够准确，则系统下发进入自动自我学习模式的指令。本申请在接收到进入自动自我学习模式的指令后，触发自动自我学习模式。
需要说明的是，标准偏误的1个标准差(1σ)偏误为31.73％，2个标准差(2σ)偏误为4.55％，2.5个标准差(2.5σ)偏误为1％，3个标准差(3σ)偏误为0.2％。由于多次获取的待校验测量值的平均值需高于多次获取的待校验测量值的中间值2.5个标准差，所以这里的预设偏误阈值选择1％。
作为一种可选的实施例，感测器的测量校正方法还包括：
预先为设备内待定义功能的第一指令地址和第二指令地址分别定义进入自我学习模式的功能和自我学习模式状态回报的功能；其中，自我学习模式包括手动自我学习模式和自动自我学习模式；
在接收到进入手动自我学习模式或自动自我学习模式的指令时，相应向第一指令地址写入表示进入手动自我学习模式的设定值或表示进入自动自我学习模式的设定值，以开始进入手动自我学习模式或自动自我学习模式；
根据当前自我学习模式的学习状态，向第二指令地址写入相应的设定值，并将第二指令地址写入相应的设定值的信息反馈至系统。
进一步地，以服务器PSU(Power Supply Unit，电源供应器)的PMBus(Power Management Bus，电源管理总线)1.2指令集为例，在PMBus1.2指令集中，D1h–D3h指令集为预留给功能扩充用，本申请可以取D1h及D2h地址来作为自我学习功能定义及扩充。
定义D1h为进入自我学习模式：若PSU PMBus指令地址D1h Bit 0写入1，则开始进入自动自我学习模式。若PSU PMBus指令地址D1h Bit 0写入0，则强迫跳出自动自我学习模式。若PSU PMBus指令地址D1h Bit 1写入1，则开始进入手动自我学习模式。若PSU PMBus指令地址D1h Bit 1写入0，则强迫跳出手动自我学习模式。
定义D2h为自我学习模式状态回报：若PSU PMBus指令地址D2h Bit 0回报1，代表自我学习完成；若PSU PMBus指令地址D2h Bit 0回报0，代表自我学习未成功，可供系统或维护人员判断是否要再进行教调。若PSU PMBus指令地址D2h Bit 1回报1，代 表正式进入自我学习模式。若PSU PMBus指令地址D2h Bit 1回报0，代表正式跳出自我学习模式，即D2h＝1代表所有模式下的自我学习模式皆成功。
则手动自我学习模式的流程为(可架设自动化测试流程，使用工业电脑来执行PMBus指令，搭配电子负载机(E-Load)完成PSU负载调整)：
工业电脑对PSU送出指令地址D1h Bit 1写入1，PSU开始进入手动自我学习模式。工业电脑等待PSU回传D2h Bit 1是否为1，如果为1，则确定PSU正式进入手动自我学习模式。
1、此时电子负载机设定为0％负载，然后等待4秒，PSU纪录感测器在0％负载时的待校验测量值。此时服务器PSU的LED(light-emitting diode，发光二极管)持续闪烁绿灯。
2、接者电子负载机设定为50％负载，然后等待4秒，PSU纪录感测器在50％负载时的待校验测量值。此时服务器PSU的LED持续闪烁绿灯。
3、最后电子负载机设定为100％负载，然后等待4秒，PSU纪录感测器在100％负载时的待校验测量值。此时服务器PSU的LED持续闪烁绿灯。
4、当服务器PSU计算完待校验测量值的数字量后，储存待校验测量值的数字量至PSU的MCU EEPROM(Electrically-Erasable Programmable Read-Only Memory，电子抹除式可复写唯读记忆体)中，作为初始设定读值，以学习到感测器初始的补偿系数，并会回传D2h＝1值给工业电脑，代表完成手动自我学习，并跳出手动自我学习模式。此时服务器PSU LED持亮绿灯。如果没回传D2h＝1值给工业电脑，代表校验未完成，会跳回流程重新校验。
服务器PSU自启动自动自我学习模式的流程为：
1、当服务器PSU比对在当前负载设定值下的感测器读值与初始设定读值，如果二者误差大于±2％，便开始执行自动自我学习模式。
2、服务器PSU会送出D2h位置等于1通知系统，以告知系统PSU正式进入自动自我学习模式。
3、此时服务器PSU会以现阶段的系统负载作为校正值，然后等待4秒，当服务器PSU计算完待校验测量值的数字量后，储存待校验测量值的数字量至PSU的MCU EEPROM中，但不会覆盖原厂设定，以学习到感测器准确的补偿系数。此时服务器PSU的LED持续亮绿灯。
4、服务器PSU会回传D2h Bit 0＝1值给系统，作为校验完毕的信号，并跳出自动自我学习模式。
服务器BMC启动服务器PSU的自动自我学习模式的流程为：
1、当服务器BMC比对在当前负载设定值下的感测器读值低于2.5个标准差，便控制服务器PSU开始执行自动自我学习模式。
2、服务器PSU通过BMC的IPMI(Intelligent Platform Management Interface，智能平台管理接口)指令送出D2h位置等于1通知系统，以告知系统PSU正式进入自动自我学习模式。
3、此时服务器PSU会以现阶段的系统负载作为校正值，然后等待4秒，当服务器PSU计算完待校验测量值的数字量后，储存待校验测量值的数字量至PSU的MCU EEPROM中，但不会覆盖原厂设定，以学习到感测器准确的补偿系数。此时服务器PSU的LED持续亮绿灯。
4、服务器PSU会回传D2h Bit 0＝1值给系统，作为校验完毕的信号，并跳出自动自我学习模式。
作为一种可选的实施例，感测器的测量校正方法还包括：
将在手动自我学习模式和自动自我学习模式下获取的待校验测量值均存储至设备内的存储器中，以通过查询存储器的存储内容确定自我学习总次数，供管理存储器的容量资源使用。
具体地，上述实施例已经提及可将在手动自我学习模式和自动自我学习模式下获取的待校验测量值均存储至设备内的存储器中，这样做的目的是：系统通过查询设备内存储器的存储内容，可确定自我学习总次数(手动自我学习次数+自动自我学习次数)，以供系统管理存储器的容量资源使用。
需要说明的是，若感测器所在的设备为服务器PSU，则上述感测器的测量校正方法均可通过服务器PSU内的MCU实现。目前，服务器PSU会由MCU来完成电源供应器中转换器开关控制、风扇控制、LED控制、监控、保护及通讯等功能，MCU就分工会分为一次侧MCU及二次侧MCU。由于二次测MCU为主要与BMC通信用的MCU，因此会将待校验测量值储存至二次测MCU的EEPROM中。服务器BMC可通过IPMI指令查询服务器PSU内MCU EEPROM的寄存器，了解服务器PSU是否进行过自我学习模式，并了解自我学习总次数。
另外，本申请采用惠式电桥平衡法，来推导及简化差动放大器的模型：
如图6所示，惠式电桥包括待测电阻Rx及已知电阻的可变电阻器R2、电阻R1和电阻R3，将R1和R2串联、R3和Rx串联，再将这两个串联的电路并联，在R1和R2之间的电线中点与在R3和Rx之间的电线中点连接上一条电线，在这条电线上放置检流计 V G。当R2/R1＝Rx/R3，检流计(电桥节点B及D)无电流通过，因此此情况下可将电桥节点B及D间的组件等效于无效并可移除。
接下将图1所示的感测器架构的等效电路图推出：
1、因分流电阻器为0.5m-1m ohm(毫欧母)，设定电阻R1、R2、R3、R4通常为数K ohm(千欧母)，所以分流电阻器可等效于短路，因此设定电阻R3和R4可对接。
2、使用克希荷夫电路定律，REF参考电压与差动放大器的输出端OUT可等效于短路，因此设定电阻R1和R2可对接。
3、推出如图7所示的感测器架构的等效电路，设定R2*R3＝R1*R4(R3/R1＝R4/R2)，采用惠式电桥平衡法，此时飘移电压等效为零。
此时差动放大器的输出为：Vo＝(R1/R3)*(V IN+-V IN-)，R1/R3为此差动放大器的增益值。
综上，得到以下结论来调整感测器硬件来达到优化：
1)通过调整R1及R3可提高差动放大器的信号放大倍率。
2)通过R2*R3＝R1*R4(R3/R1＝R4/R2)可让差动放大器的飘移电压等效为零。优化设计可让R1＝R2＝R3＝R4。
本申请还提供了提供了一种感测器的测量校正装置，该一种感测器的测量校正装置可以包括计算机设备，该计算机设备可以是终端或者服务器，该一种感测器的测量校正装置的内部结构图可以如图8所示。该一种感测器的测量校正装置包括通过系统总线连接的处理器、存储器、网络接口和输入装置。其中，该处理器用于提供计算和控制能力。该存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统和计算机可读指令。该内存储器为非易失性存储介质中的操作系统和计算机可读指令的运行提供环境。该计算机设备的网络接口用于与外部的终端或者服务器通过网络连接通信。该计算机可读指令被处理器执行时以实现一种感测器的测量校正方法。该输入装置可以是显示屏上覆盖的触摸层，也可以是计算机设备外壳上设置的按键、轨迹球或触控板，还可以是外接的键盘、触控板或鼠标等。
本领域技术人员可以理解，图8中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的设备的限定，具体的设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
本申请提供的测量校正装置的介绍请参考上述测量校正方法的实施例，本申请在此 不再赘述。
本申请实施例还提供了一种非易失性可读存储介质，该非易失性可读存储介质中存储有计算机可读指令，该计算机可读指令被一个或多个处理器执行时可实现上述任意一个实施例的感测器的测量校正方法的步骤。
本申请实施例提供的一种感测器的测量校正装置及可读存储介质中相关部分的说明可以参见本申请实施例提供的一种感测器的测量校正方法中对应部分的详细说明，在此不再赘述。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机可读指令来指令相关的硬件来完成，所述的计算机可读指令可存储于一非易失性计算机可读取存储介质中，该计算机可读指令在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用，均可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)或者外部高速缓冲存储器。作为说明而非局限，RAM以多种形式可得，诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDRSDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)等。
本申请还提供了一种服务器电源，包括感测器及上述感测器的测量校正装置；感测器包括分流电阻器及包含差动放大器、第一电阻、第二电阻、第三电阻及第四电阻的差动放大电路；其中：
差动放大器的输入负端分别与第一电阻的第一端和第三电阻的第一端连接，第一电阻的第二端与差动放大器的输出端连接，第三电阻的第二端与分流电阻器的电流输出端连接，差动放大器的输入正端分别与第二电阻的第一端和第四电阻的第一端连接，第二电阻的第二端与差动放大器的参考端连接，第四电阻的第二端与分流电阻器的电流输入端连接；
其中，第三电阻与第一电阻的比值等于第四电阻与第二电阻的比值；通过调整第一电阻和第三电阻的比值大小来调整差动放大电路的增益值。
本申请提供的服务器电源的介绍请参考上述测量校正方法及装置的实施例，本申请 在此不再赘述。
还需要说明的是，在本说明书中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
对所公开的实施例的上述说明，使本领域专业技术人员能够实现或使用本申请。对这些实施例的多种修改对本领域的专业技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本申请的精神或范围的情况下，在其他实施例中实现。因此，本申请将不会被限制于本文所示的这些实施例，而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。

Claims (11)

  1. 一种感测器的测量校正方法，其特征在于，应用于包含分流电阻器和差动放大电路的感测器，包括：
    预先对所述感测器进行补偿校验，得到所述感测器的补偿系数；
    获取所述感测器输出的电压信号；及
    根据所述分流电阻器的误差值和所述补偿系数校正所述电压信号，得到电压校正信号，以供系统优化管理使用。
  2. 如权利要求1所述的感测器的测量校正方法，其特征在于，预先对所述感测器进行补偿校验，得到所述感测器的补偿系数的过程，包括：
    在接收到进入手动自我学习模式的指令后，进入手动自我学习模式；
    根据所述感测器所在设备的负载的不同设定值，得到所述感测器在所述不同设定值下的初始待校验测量值；
    根据预设补偿系数关系式
    Figure PCTCN2021142858-appb-100001
    计算所述感测器的补偿系数；其中，S cale为所述补偿系数；
    Figure PCTCN2021142858-appb-100002
    为所述感测器对应的待测参数在第一负载设定值下的实际数字信号值；
    Figure PCTCN2021142858-appb-100003
    为所述待测参数在第二负载设定值下的实际数字信号值；
    Figure PCTCN2021142858-appb-100004
    为所述感测器在所述第一负载设定值下的初始待校验测量值对应的数字信号值；
    Figure PCTCN2021142858-appb-100005
    为所述感测器在所述第二负载设定值下的初始待校验测量值对应的数字信号值。
  3. 如权利要求2所述的感测器的测量校正方法，其特征在于，根据所述分流电阻器的误差值和所述补偿系数校正所述电压信号，得到电压校正信号的过程，包括：
    根据预设电压校正关系式
    Figure PCTCN2021142858-appb-100006
    校正所述电压信号，得到电压校正信号；
    其中，Vo_real为所述电压校正信号；ADC Count为所述电压信号对应的数字信号值；Rshunt Count为所述分流电阻器的误差值对应的数字信号值；V Gain为所述差动放大电路的增益值。
  4. 如权利要求2或3所述的感测器的测量校正方法，其特征在于，所述感测器的测量校正方法还包括：
    在触发自动自我学习模式时，进入自动自我学习模式；
    获取所述感测器在当前负载设定值下的新待校验测量值；及
    根据预设补偿系数更新关系式
    Figure PCTCN2021142858-appb-100007
    计算所述感测器更新的补偿系数，以根据所述更新的补偿系数校正所述电压信号；其中，S′ cale为所述更新的补偿系数；
    Figure PCTCN2021142858-appb-100008
    为所述新待校验测量值对应的数字信号值；
    Figure PCTCN2021142858-appb-100009
    为所述待测参数在当前负载设定值下的实际数字信号值。
  5. 如权利要求4所述的感测器的测量校正方法，其特征在于，触发自动自我学习模式的过程，包括：
    在所述感测器所在设备运行时，多次获取所述感测器在当前负载设定值下的待校验测量值，并根据移动平均算法求取获取的多个待校验测量值的平均待校验测量值；及
    在所述平均待校验测量值与所述感测器在当前负载设定值下对应的初始待校验测量值的差值大于预设误差阈值时，触发自动自我学习模式。
  6. 如权利要求4所述的感测器的测量校正方法，其特征在于，触发自动自我学习模式的过程，包括：
    在接收到进入自动自我学习模式的指令后，触发自动自我学习模式；
    其中，所述进入自动自我学习模式的指令的下发条件为：多次获取的所述感测器在当前负载设定值下的待校验测量值的平均值高于多次获取的待校验测量值的中间值的差值未达到预设偏误阈值。
  7. 如权利要求4所述的感测器的测量校正方法，其特征在于，所述感测器的测量校正方法还包括：
    预先为所述设备内待定义功能的第一指令地址和第二指令地址分别定义进入自我学习模式的功能和自我学习模式状态回报的功能；其中，所述自我学习模式包括手动自我学习模式和自动自我学习模式；
    在接收到进入手动自我学习模式或自动自我学习模式的指令时，相应向所述第一指令地址写入表示进入手动自我学习模式的设定值或表示进入自动自我学习模式的设定值，以开始进入手动自我学习模式或自动自我学习模式；及
    根据当前自我学习模式的学习状态，向所述第二指令地址写入相应的设定值，并将所述第二指令地址写入所述相应的设定值的信息反馈至所述系统。
  8. 如权利要求4所述的感测器的测量校正方法，其特征在于，所述感测器的测量校正方法还包括：
    将在所述手动自我学习模式和所述自动自我学习模式下获取的待校验测量值均存储至所述设备内的存储器中，以通过查询所述存储器的存储内容确定自我学习总次数，供管理所述存储器的容量资源使用。
  9. 一种感测器的测量校正装置，其特征在于，包括存储器及一个或多个处理器，所述存储器中储存有计算机可读指令，所述计算机可读指令被所述一个或多个处理器执行时，使得所述一个或多个处理器执行如权利要求1-8任意一项所述的方法的步骤。
  10. 一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，其特征在于，所述计算机可读指令被一个或多个处理器执行时，使得所述一个或多个处理器执行如权利要求1-8任意一项所述的方法的步骤。
  11. 一种服务器电源，其特征在于，包括感测器及如权利要求9所述的感测器的测量校正装置；所述感测器包括分流电阻器及包含差动放大器、第一电阻、第二电阻、第三电阻及第四电阻的差动放大电路；其中：
    所述差动放大器的输入负端分别与所述第一电阻的第一端和所述第三电阻的第一端连接，所述第一电阻的第二端与所述差动放大器的输出端连接，所述第三电阻的第二端与所述分流电阻器的电流输出端连接，所述差动放大器的输入正端分别与所述第二电阻的第一端和所述第四电阻的第一端连接，所述第二电阻的第二端与所述差动放大器的参考端连接，所述第四电阻的第二端与所述分流电阻器的电流输入端连接；
    其中，所述第三电阻与所述第一电阻的比值等于所述第四电阻与所述第二电阻的比值；通过调整所述第一电阻和所述第三电阻的比值大小来调整所述差动放大电路的增益值。
PCT/CN2021/142858 2021-08-05 2021-12-30 一种感测器的测量校正方法、装置及服务器电源 WO2023010790A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/259,735 US20240061061A1 (en) 2021-08-05 2021-12-30 Measurement correction method and apparatus for sensor, and server power supply

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110893892.2 2021-08-05
CN202110893892.2A CN113341362B (zh) 2021-08-05 2021-08-05 一种感测器的测量校正方法、装置及服务器电源

Publications (1)

Publication Number Publication Date
WO2023010790A1 true WO2023010790A1 (zh) 2023-02-09

Family

ID=77480719

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/142858 WO2023010790A1 (zh) 2021-08-05 2021-12-30 一种感测器的测量校正方法、装置及服务器电源

Country Status (3)

Country Link
US (1) US20240061061A1 (zh)
CN (1) CN113341362B (zh)
WO (1) WO2023010790A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117269879A (zh) * 2023-09-28 2023-12-22 江苏森维电子有限公司 一种智能电表累计误差消除方法

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113341362B (zh) * 2021-08-05 2021-10-15 苏州浪潮智能科技有限公司 一种感测器的测量校正方法、装置及服务器电源
TWI822552B (zh) * 2023-01-10 2023-11-11 尚偉機電有限公司 用電偵測分配系統

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TW200834049A (en) * 2007-02-13 2008-08-16 Li-Cheng Liu An automatic calibrated measuring method for pressure
CN102269776A (zh) * 2010-06-04 2011-12-07 凌力尔特有限公司 电流检测电阻器中老化漂移的动态补偿
US20140316735A1 (en) * 2013-04-22 2014-10-23 Richtek Technology Corporation Protection Device and Calibration Method Thereof
JP2015224975A (ja) * 2014-05-28 2015-12-14 カルソニックカンセイ株式会社 バッテリ充放電電流検出装置
CN105659326A (zh) * 2013-07-16 2016-06-08 海拉企业中心(美国)公司 具有偏移校准的电流感测电路
CN106405460A (zh) * 2016-07-06 2017-02-15 广州维思车用部件有限公司 电子仪表电压检测校准系统及校准方法
JP2017129526A (ja) * 2016-01-22 2017-07-27 株式会社デンソー 電流補正回路
CN108089141A (zh) * 2017-11-16 2018-05-29 山东联合电力技术有限公司 一种基于分流器的电流测量装置的误差修正方法及装置
CN109870666A (zh) * 2017-12-04 2019-06-11 北京长城华冠汽车科技股份有限公司 电流检测校准方法
CN112578328A (zh) * 2020-12-10 2021-03-30 苏州浪潮智能科技有限公司 一种电流侦测校正电路及方法
CN113341362A (zh) * 2021-08-05 2021-09-03 苏州浪潮智能科技有限公司 一种感测器的测量校正方法、装置及服务器电源

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8930152B2 (en) * 2009-09-25 2015-01-06 University Of Washington Whole structure contactless power consumption sensing
CN105353335B (zh) * 2015-11-23 2018-01-05 国家电网公司 一种交流分压器自动校验装置及方法
CN106093521A (zh) * 2016-06-17 2016-11-09 广州极飞电子科技有限公司 电流检测方法、装置及电池管理系统
CN106226682B (zh) * 2016-08-01 2019-05-31 广东美的制冷设备有限公司 功率因数校正器及其电流检测电路的故障诊断方法、装置
CN106950524B (zh) * 2017-02-22 2019-08-20 歌尔科技有限公司 用于电压测量装置的校准方法、装置及校准系统
CN108663558B (zh) * 2017-03-29 2020-06-23 株式会社村田制作所 一种pfc设备及其电流检测方法和电流检测装置
CN112557987A (zh) * 2020-12-18 2021-03-26 珠海市运泰利自动化设备有限公司 一种电流测量校准系统及方法

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TW200834049A (en) * 2007-02-13 2008-08-16 Li-Cheng Liu An automatic calibrated measuring method for pressure
CN102269776A (zh) * 2010-06-04 2011-12-07 凌力尔特有限公司 电流检测电阻器中老化漂移的动态补偿
US20140316735A1 (en) * 2013-04-22 2014-10-23 Richtek Technology Corporation Protection Device and Calibration Method Thereof
CN105659326A (zh) * 2013-07-16 2016-06-08 海拉企业中心(美国)公司 具有偏移校准的电流感测电路
JP2015224975A (ja) * 2014-05-28 2015-12-14 カルソニックカンセイ株式会社 バッテリ充放電電流検出装置
JP2017129526A (ja) * 2016-01-22 2017-07-27 株式会社デンソー 電流補正回路
CN106405460A (zh) * 2016-07-06 2017-02-15 广州维思车用部件有限公司 电子仪表电压检测校准系统及校准方法
CN108089141A (zh) * 2017-11-16 2018-05-29 山东联合电力技术有限公司 一种基于分流器的电流测量装置的误差修正方法及装置
CN109870666A (zh) * 2017-12-04 2019-06-11 北京长城华冠汽车科技股份有限公司 电流检测校准方法
CN112578328A (zh) * 2020-12-10 2021-03-30 苏州浪潮智能科技有限公司 一种电流侦测校正电路及方法
CN113341362A (zh) * 2021-08-05 2021-09-03 苏州浪潮智能科技有限公司 一种感测器的测量校正方法、装置及服务器电源

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
FANG WENTIAN: "Design of Correction Device for Harmonic Measurement of Capacitor Voltage Transformer", HEILONGJIANG ELECTRIC POWER, vol. 6, no. 3, 15 June 2020 (2020-06-15), pages 208 - 214, XP093033424, ISSN: 2095-6843, DOI: 10.13625/j.cnki.hljep.2020.03.005 *
WANG HU;YING ZHONG-DE;CHEN MING-LIANG: "PMSM Current Sampling Error Elimination Method Based on Voltage Feedback", CONTROL ENGINEERING OF CHINA, vol. 28, no. 7, 20 July 2021 (2021-07-20), pages 1321 - 1327, XP093033420, ISSN: 1671-7848, DOI: 10.14107/j.cnki.kzgc.20190451 *
WU HAONAN; ZHAO MENGLIAN; YANG ZHAO; LIU SHENG; WU XIAOBO: "Design of a Low Temperature Drift UVLO Circuit with Base Current Compensation", 2019 IEEE INTERNATIONAL CONFERENCE ON ELECTRON DEVICES AND SOLID-STATE CIRCUITS (EDSSC), 12 June 2019 (2019-06-12), pages 1 - 3, XP033573047, DOI: 10.1109/EDSSC.2019.8754320 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117269879A (zh) * 2023-09-28 2023-12-22 江苏森维电子有限公司 一种智能电表累计误差消除方法
CN117269879B (zh) * 2023-09-28 2024-03-29 江苏森维电子有限公司 一种智能电表累计误差消除方法

Also Published As

Publication number Publication date
US20240061061A1 (en) 2024-02-22
CN113341362B (zh) 2021-10-15
CN113341362A (zh) 2021-09-03

Similar Documents

Publication Publication Date Title
WO2023010790A1 (zh) 一种感测器的测量校正方法、装置及服务器电源
WO2020253175A1 (zh) 一种甲醛浓度检测方法、装置及空气净化器
EP2847562B1 (en) Methods and apparatus to detect leakage current in a resistance temperature detector
TWI653805B (zh) 用於校準基於庫侖計數的充電狀態估計的方法和裝置
CN109470407B (zh) 分布式多节点液体温度、压力传感器测量数据的校准方法
TWI535136B (zh) 保護裝置及其校正方法
CN101706294B (zh) 一种自动判断传感器的校准时间的方法
CN114047472B (zh) 一种智能电表的计量误差监测系统及其监测方法、装置
CN111157081B (zh) 一种电子式燃气表的校准方法
CN111998918A (zh) 一种误差校正方法、误差校正装置及流量传感系统
CN112799493B (zh) 一种电源vr芯片的电流自动校准电路及校准方法
KR101018702B1 (ko) 자기교정과 추적관리 기능을 갖는 부분방전량 교정기
CN113625215B (zh) 基于分段测试的电压互感器异常校准方法及装置
WO2016010920A1 (en) Smart meter system architecture
US5621329A (en) Automatic self-calibration system for digital teraohmmeter
CN108919063B (zh) 一种基于电容修正原理的电场遥测系统及方法
WO2023173873A1 (zh) 一种校正方法及相关组件
CN109407042B (zh) 一种智能电表的校验方法
TWI721532B (zh) 電路板故障診斷裝置及診斷方法
CN114518185A (zh) 热电阻温度传感器的温度校验方法、系统、计算机及介质
KR20040076705A (ko) 손실 계수의 오차 보정을 이용한 게이트 커패시턴스 측정방법
CN115016437A (zh) 一种伺服系统产品位置标定装置及方法
CN104122917B (zh) 保护装置及其校正方法
US20220350352A1 (en) Power source with error detection
TWI614510B (zh) 濕度傳感器的校正方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21952653

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 18259735

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21952653

Country of ref document: EP

Kind code of ref document: A1