An automatic time calibration system for power plant and station electric energy meters

Abstract

The invention discloses an automatic time calibration system for electric energy meters in a plant or station, comprising a calibration information acquisition unit, a protocol identification and judgment unit, a protocol comprehensive selection unit, a calibration error analysis unit, and a time calibration information output unit. The invention relates to the technical field of electric energy meter calibration, and solves the technical problem of reducing the influence of hardware and software errors and improving the accuracy of the overall time calibration error. The invention analyzes the compatibility of electric energy meter protocols of equipment, judges the compatibility result, analyzes different electric energy meter protocols for the case of incompatible results, and selects a standard electric energy meter protocol based on the corresponding transmission rate and delay, then analyzes from two aspects of hardware error and software error respectively, performs corresponding error calculation according to the analysis result, and performs time calibration based on the obtained error calculation result, and improves the accuracy of the overall time calibration by analyzing the influence of both hardware error and software error.

Description

Technical Field

[0001] The invention relates to the technical field of electric energy meter calibration, and in particular to an automatic time calibration system for electric energy meters in power plants. Background Art

[0002] As a key metering device in the power system, the accuracy of the energy meter directly affects the fairness of electricity bill settlement, the rationality of energy allocation and the stability of system operation. However, due to errors in the manufacturing process, wear and tear caused by long-term use, and the influence of environmental factors such as temperature and humidity changes, the measurement accuracy of the energy meter may gradually decrease and errors may occur.

[0003] According to the publication number CN118534405A, a calibration method for an electric energy meter is disclosed, which relates to the technical field of electric energy meter calibration, including obtaining the historical installation and maintenance records of all electric energy meters, and calculating the deviation correlation coefficient; obtaining the daily deviation of each maintenance cycle, and calculating the real-time calibration cycle of the electric energy meter; obtaining the usage time of a new meter without a maintenance record and the total measured electricity of the new meter, and calculating the real-time error rate of the new meter in combination with the deviation correlation coefficient; extracting the electric energy meters that have been automatically calibrated in the power grid system for standard meter error analysis, and obtaining the automatic calibration evaluation index after calculation, judging the automatic calibration accuracy of the electric energy meter in the current random inspection cycle according to the automatic calibration evaluation index, and selecting different calibration warning processing strategies.

[0004] When some existing time calibration systems are in use, due to the influence of hardware and software during the time calibration, if the corresponding errors are not eliminated, the subsequent time calibration operations will affect the overall time calibration accuracy. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides an automatic time calibration system for power plant electricity meters, which solves the problem of reducing the impact of hardware and software errors and improving the accuracy of the overall time calibration error.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an automatic time calibration system for power station electric energy meters, comprising:

[0007] A calibration information acquisition unit, used to acquire the time information and meter protocol in the calibration of the electric energy meter, and transmit the acquired time information and meter protocol to the protocol identification and judgment unit;

[0008] A protocol identification and judgment unit, used for identifying and judging the meter protocol transmitted by the correction information acquisition unit, generating a protocol judgment result by judging the compatibility of the meter protocol, and the protocol judgment result includes a compatibility result and an incompatibility result, and transmitting the compatibility result to the correction error analysis unit, and transmitting the incompatibility result to the protocol comprehensive selection unit;

[0009] A protocol comprehensive selection unit is used to analyze the meter protocol corresponding to the incompatibility result, calculate the comprehensive value corresponding to the meter protocol, select the standard meter protocol according to the comprehensive value and equipment requirements, and then transmit the standard meter protocol to the correction error analysis unit;

[0010] The correction error analysis unit is used to analyze the correction records according to the standard meter protocol, and to calculate the impact in combination with the hardware error and the software error. For the hardware error, the corresponding impact factor is determined by analyzing the impact of temperature. For the software error, the clock drift compensation is calculated by the Kalman filter algorithm. The time correction analysis is performed in combination with the hardware error and the software error to generate the time correction information, and the time correction information is transmitted to the time correction information output unit.

[0011] The time calibration information output unit is used to display the generated time calibration information to the corresponding operator.

[0012] As a further solution of the present invention: the specific manner in which the protocol identification and judgment unit judges the meter protocol and generates the protocol judgment result is:

[0013] Two sets of electricity meter protocols are obtained, and the message headers and message tails in the electricity meter protocols are identified and judged. If both protocols use similar message headers to identify the beginning of the message and use message tails to identify the end of the message, it indicates that the two are compatible and a compatibility result is generated. Otherwise, if either the message header or the message tail is different, it indicates that there is no compatibility and an incompatible result is generated.

[0014] As a further solution of the present invention: the protocol comprehensive selection unit analyzes the meter protocol corresponding to the compatibility result and selects the standard meter protocol in the following manner:

[0015] All meter protocols are obtained and labeled as i, where i = 1, 2, ..., j, where j represents the number of meter protocols. Then the transmission rate Vi and transmission delay Si corresponding to meter protocol i are obtained, and the obtained transmission rate Vi and transmission delay Si are substituted into the formula The comprehensive value Wi corresponding to the meter protocol i is calculated, and the meter protocols i are sorted from large to small according to the calculated comprehensive value Wi. Then, the equipment requirements are obtained, and the meter protocol corresponding to the largest comprehensive value Wi is selected based on the equipment requirements, and standard protocol information is generated.

[0016] As a further solution of the present invention: the correction error analysis unit determines the corresponding influencing factor by analyzing the influence of temperature on the hardware error in a specific manner:

[0017] Obtain multiple calibration records and label them as n, where n=1, 2, ..., m, where m represents the number of calibration records, obtain the real-time temperature of the crystal oscillator corresponding to calibration record n ​​as Gn, and obtain the temperature corresponding to the normal operation of the crystal oscillator as Gz, and then identify and classify the real-time temperature Gn in the calibration record according to the normal temperature Gz;

[0018] The correction records corresponding to the case where the real-time temperature Gn is higher than the normal temperature Gz are classified as high-temperature correction records, otherwise they are classified as low-temperature correction records. At the same time, the influencing change factors corresponding to the high-temperature correction records and the low-temperature correction records are calculated respectively.

[0019] As a further solution of the present invention: the specific manner in which the correction error analysis unit analyzes the high temperature correction record and the low temperature correction record separately is:

[0020] The calculation of the high temperature calibration record influence change factor is as follows: the high temperature calibration records are sorted from low to high according to the corresponding real-time temperature, and the calibration errors corresponding to different real-time temperatures are obtained at the same time, and then a relationship diagram between temperature and calibration error is established, and the corresponding high temperature influence change factor under high temperature conditions is obtained according to the relationship diagram;

[0021] The calculation of the low temperature correction record influence change factor is as follows: all low temperature correction records are sorted from low to high according to the corresponding real-time temperature, and the corresponding low temperature influence change factor is obtained in the same way as the high temperature correction record analysis method.

[0022] As a further solution of the present invention: the correction error analysis unit calculates the clock drift compensation for the software error through the Kalman filter algorithm in the following specific manner:

[0023] In the context of clock compensation, the clock offset and the clock drift rate are usually taken as elements of the state vector. Let the state vector x = [θ, w] T , where θ represents the clock offset (the difference from the standard time), W represents the clock drift rate, and the state transfer equation is established based on the physical characteristics of the clock;

[0024] In discrete time, the clock offset at the next time k+1 is the offset at the current time k plus the drift rate multiplied by the time interval Δt, so the state transfer equation can be expressed as x k+1 =Fx k +w k ,in w k is the process noise, which is used to describe the uncertainty of the model;

[0025] Assume the measured value is zk, the measured value is usually the observed value of the clock deviation obtained by comparing it with a standard clock source. The measurement equation is to link the measured value with the state vector, which is generally expressed as z k =Hx k +v1, where H = [1, 0], V k is the measurement noise, which reflects the uncertainty in the measurement process;

[0026] Next, the initial state is estimated. During the estimation, the clock offset and drift rate need to be initially estimated and determined by initial measurements when the system starts running.

[0027] At the same time, the initial covariance matrix is ​​estimated, and the covariance matrix P0 of the initial state is estimated. When the initial estimate is uncertain, a larger covariance value is set. When the initial estimate is certain, a smaller covariance value is set, and the Kalman filter algorithm is used for iterative calculation.

[0028] As a further solution of the present invention: the specific method in which the correction error analysis unit uses the Kalman filter algorithm to perform iterative calculation is:

[0029] For the prediction stage: Calculate the predicted value of the state according to the state transfer equation At the same time, calculate the prediction covariance matrix P k|k-1 =FP k-1|k-1 F T +Q, where Q is the process noise covariance matrix, which specifically reflects the intensity of the process noise;

[0030] For the update phase: calculate the Kalman gain K k =P K|K-1 H T (HP K|K-1 H T +R) -1 , where R is the measurement noise covariance matrix, and then the state estimate is updated according to the measurement value Finally update the covariance matrix P k|k =(IK k H)P k|k-1 , where I represents the identity matrix, and (IK k H) is the Kalman gain K k The product of the measurement matrix H is corrected;

[0031] For the clock compensation phase: the optimal estimate of the clock offset obtained at each iteration The clock is compensated by adjusting the clock time. The corresponding compensation amount is

[0032] The hardware error impact factor and the software error impact are summed up to obtain a comprehensive error impact value. The comprehensive error impact value is used as a standard for time calibration and time calibration information is generated.

[0033] The present invention provides an automatic time calibration system for power plant electric energy meters. Compared with the prior art, it has the following advantages:

[0034] Beneficial effects:

[0035] The present invention analyzes the compatibility of the meter protocol of the device to determine the compatibility result. For the case of incompatible results, different meter protocols are analyzed, and a standard meter protocol is selected based on the corresponding transmission rate and delay. Secondly, analysis is performed from two aspects: hardware error and software error. According to the results of the analysis, corresponding error calculation is performed, and time calibration is performed based on the obtained error calculation result. By analyzing the impact of both hardware error and software error, the overall time calibration accuracy is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a block diagram of the system principle of the present invention;

[0037] Figure 2 This is the schematic diagram of the time synchronization host system;

[0038] Figure 3 This is the schematic diagram of the time synchronization slave system. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] For example, see Figure 1 and Figure 3 The present application provides an automatic time calibration system for power station electric energy meters, including: a correction information acquisition unit, a protocol identification and judgment unit, a protocol comprehensive selection unit, a correction error analysis unit, and a time calibration information output unit, and the correction information acquisition unit, the protocol identification and judgment unit, the correction error analysis unit and the time calibration information output unit are unidirectionally electrically connected.

[0041] The correction information acquisition unit is used to acquire the time information and meter protocol in the time calibration of the electric energy meter, and transmit the acquired time information and meter protocol to the protocol identification and judgment unit. The specific time information is mainly received by the timing host through the Beidou time information, and the acquired time information is transcoded at the same time, while the meter protocol is acquired through the timing slave.

[0042] The protocol identification and judgment unit is used to identify and judge the acquired meter protocol and generate a protocol judgment result. The specific identification and judgment method is:

[0043] According to the meter protocol obtained from the synchronization slave, the compatibility of the meter protocol is determined at the same time. The specific method for determining the compatibility is as follows: two sets of meter protocols are obtained to check whether the protocols contain specific message header and message tail identifiers. If both protocols use similar message headers to identify the beginning of a message and message tails to identify the end of a message, this may indicate that they have a certain degree of compatibility in message encapsulation and generate a compatibility result. Otherwise, if any of the message headers or message tails are different, this indicates that they are not compatible and generates an incompatible result.

[0044] In a specific example, for example, there are protocols A and B. Protocol A's messages always start with a specific character sequence "[START]" as the message header and end with " " as the message footer; Protocol B also uses "[START]" as the message header and " " as the message footer, so the two protocols are compatible in terms of message encapsulation. If, however, Protocol C uses "[BEGIN]" as the message header and " " as the message footer, the message header is different from that of Protocol A, and it can be determined that Protocol A and Protocol C are not compatible.

[0045] For the generated compatibility result, the corresponding meter protocol is transmitted to the correction error analysis unit, and for the generated incompatibility result, it is transmitted to the protocol comprehensive selection unit.

[0046] The protocol comprehensive selection unit is used to analyze the meter protocol according to the generated incompatibility result, and select the standard meter protocol by calculating the comprehensive value of the meter protocol, and the calculation method of the comprehensive value is:

[0047] All meter protocols are obtained and labeled as i, where i = 1, 2, ..., j, where j represents the number of meter protocols. The meter protocols obtained here are the meter protocols that can be recognized by the timing host and the timing slave at the same time. Then, the transmission rate Vi and transmission delay Si corresponding to the meter protocol i are obtained, and the obtained transmission rate Vi and transmission delay Si are substituted into the formula The comprehensive value Wi corresponding to the meter protocol i is calculated, and the meter protocols i are sorted from large to small according to the calculated comprehensive value Wi. Then, the equipment requirements are obtained, and the equipment requirements here refer to the actual usage requirements. The meter protocol corresponding to the largest comprehensive value Wi is selected in combination with the equipment requirements, and the standard protocol information is generated. At the same time, the standard protocol information is transmitted to the correction error analysis unit.

[0048] Embodiment 2, as embodiment 2 of the present invention, differs from embodiment 1 in that the correction error analysis unit performs timing analysis according to the acquired standard protocol information.

[0049] The calibration error analysis unit acquires and analyzes multiple calibration records and performs comprehensive processing from both hardware error and software error aspects.

[0050] The specific method for hardware error analysis is to adjust the crystal oscillator frequency and calculate the influencing factors under different temperature conditions to analyze and adjust. The specific analysis method is:

[0051] Obtain multiple calibration records and label them as n, where n=1, 2, ..., m, where m represents the number of calibration records, and obtain the real-time temperature of the crystal oscillator corresponding to calibration record n ​​as Gn, and the temperature here is obtained by real-time monitoring by a built-in temperature sensor, and at the same time obtain the temperature corresponding to the normal working state of the crystal oscillator as Gz, and then identify and classify the real-time temperature Gn in the calibration record according to the normal temperature Gz;

[0052] In a specific example, for example, assume that there are 5 calibration records (i.e., m=5), which are numbered 1, 2, 3, 4, and 5. The temperature sensor measures that the real-time temperature G1 of the crystal oscillator corresponding to calibration record 1 is 30°C, the real-time temperature G2 corresponding to calibration record 2 is 32°C, the real-time temperature G3 corresponding to calibration record 3 is 28°C, the real-time temperature G4 corresponding to calibration record 4 is 35°C, and the real-time temperature G5 corresponding to calibration record 5 is 25°C. Assume that the temperature G corresponding to the normal operation of the crystal oscillator is 30°C±2°C.

[0053] The correction records corresponding to the case where the real-time temperature Gn is higher than the normal temperature Gz are classified as high-temperature correction records, and the corresponding symbol for "higher" is "≥", and the correction records corresponding to the case where the real-time temperature Gn is lower than the normal temperature Gz are classified as low-temperature correction records, and the corresponding symbol for "lower" is "<". At the same time, the influencing change factors corresponding to the high-temperature correction records and the low-temperature correction records are calculated respectively; then based on the numerical values ​​in the above example, G1, G2, and G3 can be classified as low-temperature correction records within the normal temperature range, and G4 and G5 can be classified as high-temperature correction records.

[0054] The calculation of the change factor affecting the high-temperature correction record is as follows: sort the high-temperature correction records from low to high according to the corresponding real-time temperature, obtain the correction error corresponding to different real-time temperatures, then establish a relationship diagram between temperature and correction error, and obtain the corresponding high-temperature impact change factor under high temperature conditions based on the relationship diagram.

[0055] In a specific example, there is a set of high-temperature calibration records, and the real-time temperatures are 50°C, 55°C, 60°C, 65°C and 70°C. After measurement, the calibration errors corresponding to these temperatures are 0.1, 0.15, 0.22, 0.3 and 0.4 respectively. After sorting the temperatures from low to high, a relationship diagram is drawn with temperature as the horizontal axis and calibration error as the vertical axis. In the relationship diagram, it can be seen that as the temperature increases, the calibration error shows a certain trend of change. Based on this trend, the influencing change factor under high temperature conditions (for example, more than 60°C) can be determined. If it is found from the figure that the calibration error increases by approximately 0.08 for every 5°C increase in temperature, then this rule can be used as a high temperature influencing change factor for subsequent analysis or calculation under high temperature conditions.

[0056] The calculation of the low temperature correction record influence change factor is as follows: all low temperature correction records are sorted from low to high according to the corresponding real-time temperature, and the corresponding low temperature influence change factor is obtained in the same way as the high temperature correction record analysis method.

[0057] The specific method for software error analysis is to analyze through the corresponding clock drift compensation: In the context of clock compensation, the clock offset and the clock drift rate are usually used as elements of the state vector. For example, let the state vector x = [θ, w] T , where θ represents the clock offset (the difference from the standard time), W represents the clock drift rate, and the state transfer equation is established according to the physical characteristics of the clock, and the specific state transfer equation is as follows:

[0058] Assuming that in discrete time, the clock offset at the next time k+1 is the offset at the current time k plus the drift rate multiplied by the time interval Δt, and the drift rate remains unchanged in ideal conditions (it may change slowly in practice), then the state transfer equation can be expressed as x k+1 =Fx k +w k ,in w k is the process noise, which is used to describe the uncertainty of the model.

[0059] Let the measured value be z k The measured value is usually the observed value of the clock deviation obtained by comparing it with a standard clock source (such as GPS clock, atomic clock, etc.). The measurement equation is to link the measured value with the state vector, which is generally expressed as z k =Hxk +v k , where H = [1, 0], because the clock offset is directly measured, V k is the measurement noise, which reflects the uncertainty in the measurement process.

[0060] Next, the initial state is estimated. When estimating, the clock offset and drift rate need to be initially estimated. This can be determined by the initial measurement when the system starts running. For example, assuming the initial clock offset is The initial drift rate is the value obtained by comparing with the standard clock for the first time. It can be estimated based on experience or multiple measurements over a short period of time.

[0061] At the same time, the initial covariance matrix is ​​estimated. During the estimation process, the covariance matrix P0 of the initial state needs to be estimated, which describes the uncertainty of the initial state estimation. For the case where the initial estimate is relatively uncertain, a larger covariance value can be set, and for the case where the initial estimate is relatively certain, a smaller covariance value is set.

[0062] Then the Kalman filter algorithm is used for iterative calculation, and the specific iterative calculation process is as follows:

[0063] For the prediction stage: Calculate the predicted value of the state according to the state transfer equation At the same time, calculate the prediction covariance matrix P k|k-1 =FP k-1|k-1 F T +Q, where Q is the process noise covariance matrix, which specifically reflects the intensity of the process noise.

[0064] For the update phase: calculate the Kalman gain K k =P K|K-1 H T (HP K|K-1 H T +R) -1 , where R is the measurement noise covariance matrix, and then the state estimate is updated according to the measurement value Finally update the covariance matrix P k|k =(IK k H)P k|k-1 , where l represents the identity matrix. The specific identity matrix is ​​a square matrix whose main diagonal elements are all 1 and the remaining elements are all 0. At the same time, (IK k H) is the Kalman gain K k The correction is performed by multiplying the measurement matrix H.

[0065] For the clock compensation phase: the optimal estimate of the clock offset obtained at each iteration The clock is compensated, and the specific compensation method here is to adjust the clock time to make it closer to the standard time. The corresponding compensation amount is For example, if the current clock shows time t c , the standard time should be Then adjust the clock to

[0066] The hardware error influence factor and the software error influence are summed up to obtain a comprehensive error influence value. At the same time, the time calibration is performed based on the comprehensive error influence value, and the time calibration information is generated, which is then transmitted to the time calibration information output unit.

[0067] The time calibration information output unit is used to display the generated time calibration information to the corresponding operator.

[0068] Embodiment 3, as the embodiment 3 of the present invention, focuses on combining the implementation processes of embodiment 1 and embodiment 2 for implementation.

[0069] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0070] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An automatic time calibration system for electric energy meters in power plants, characterized in that: include: A protocol identification and judgment unit, used for identifying and judging the meter protocol transmitted by the correction information acquisition unit, generating a protocol judgment result by judging the compatibility of the meter protocol, and the protocol judgment result includes a compatibility result and an incompatibility result, and transmitting the compatibility result to the correction error analysis unit, and transmitting the incompatibility result to the protocol comprehensive selection unit; The protocol comprehensive selection unit is used to analyze the meter protocol corresponding to the incompatibility result and calculate the comprehensive value corresponding to the meter protocol. At the same time, the standard meter protocol is selected according to the comprehensive value and equipment requirements, and then the standard meter protocol is transmitted to the correction error analysis unit. The specific analysis method is as follows: All meter protocols are obtained and labeled as i, where i = 1, 2, ..., j, where j represents the number of meter protocols. Then the transmission rate Vi and transmission delay Si corresponding to meter protocol i are obtained, and the obtained transmission rate Vi and transmission delay Si are substituted into the formula Calculate the comprehensive value Wi corresponding to the meter protocol i, and sort the meter protocols i from large to small according to the calculated comprehensive value Wi, then obtain the equipment requirements, and select the meter protocol corresponding to the maximum comprehensive value Wi based on the equipment requirements, and generate standard protocol information; The correction error analysis unit is used to analyze the correction records according to the standard meter protocol, and to calculate the impact in combination with the hardware error and the software error. For the hardware error, the corresponding impact factor is determined by analyzing the impact of temperature. For the software error, the clock drift compensation is calculated by the Kalman filter algorithm. The time correction analysis is performed in combination with the hardware error and the software error to generate the time correction information, and the time correction information is transmitted to the time correction information output unit.

2. The automatic time calibration system for power plant electric energy meters according to claim 1 is characterized in that: The automatic time calibration system also includes a calibration information acquisition unit and a time calibration information output unit; A calibration information acquisition unit, used to acquire the time information and meter protocol in the calibration of the electric energy meter, and transmit the acquired time information and meter protocol to the protocol identification and judgment unit; The time calibration information output unit is used to display the generated time calibration information to the corresponding operator.

3. The automatic time calibration system for power plant electric energy meters according to claim 1 is characterized in that: The specific method in which the protocol identification and judgment unit judges the meter protocol and generates the protocol judgment result is: Two sets of electricity meter protocols are obtained, and the message headers and message tails in the electricity meter protocols are identified and judged. If both protocols use the same message header to identify the beginning of the message and the message tail to identify the end of the message, it indicates that the two are compatible and a compatibility result is generated. Otherwise, if either the message header or the message tail is different, it indicates that there is no compatibility and an incompatible result is generated.

4. The automatic time calibration system for power plant electric energy meters according to claim 1 is characterized in that: The specific method in which the correction error analysis unit determines the corresponding influencing factor by analyzing the influence of temperature on the hardware error is: Obtain multiple calibration records and label them as n, where n=1, 2, ..., m, where m represents the number of calibration records, obtain the real-time temperature of the crystal oscillator corresponding to calibration record n ​​as Gn, and obtain the temperature corresponding to the normal operation of the crystal oscillator as Gz, and then identify and classify the real-time temperature Gn in the calibration record according to the normal temperature Gz; The correction records corresponding to the case where the real-time temperature Gn is higher than the normal temperature Gz are classified as high-temperature correction records, otherwise they are classified as low-temperature correction records. At the same time, the influencing change factors corresponding to the high-temperature correction records and the low-temperature correction records are calculated respectively.

5. The automatic time calibration system for power plant electric energy meters according to claim 4 is characterized in that: The specific manner in which the correction error analysis unit analyzes the high temperature correction record and the low temperature correction record separately is: The calculation of the high temperature calibration record influence change factor is as follows: the high temperature calibration records are sorted from low to high according to the corresponding real-time temperature, and the calibration errors corresponding to different real-time temperatures are obtained at the same time, and then a relationship diagram between temperature and calibration error is established, and the corresponding high temperature influence change factor under high temperature conditions is obtained according to the relationship diagram; The calculation of the low temperature correction record influence change factor is as follows: all low temperature correction records are sorted from low to high according to the corresponding real-time temperature, and the corresponding low temperature influence change factor is obtained in the same way as the high temperature correction record analysis method.

6. The automatic time calibration system for power plant electric energy meters according to claim 1 is characterized in that: The specific method of calculating the clock drift compensation of the software error by the correction error analysis unit through the Kalman filter algorithm is as follows: In the context of clock compensation, the clock offset and the clock drift rate are taken as elements of the state vector, and the state vector x = [θ, w] T , where θ represents the offset of the clock, W represents the drift rate of the clock, and the state transfer equation is established according to the physical characteristics of the clock; In discrete time, the clock offset at the next time k+1 is the offset at the current time k plus the drift rate multiplied by the time interval Δt, so the state transfer equation can be expressed as x k+1 =Fx k +w k ,in w k is the process noise, which is used to describe the uncertainty of the model; Let the measured value be Z k , and the measurement value is linked to the state vector to obtain the measurement equation, expressed as z k =Hx k +v k , where H = [1,0], V k is the measurement noise, which reflects the uncertainty in the measurement process; Next, the initial state is estimated. During the estimation, the clock offset and drift rate need to be initially estimated and determined by initial measurements when the system starts running. At the same time, the initial covariance matrix is ​​estimated, and the covariance matrix P0 of the initial state is estimated. When the initial estimate is uncertain, the maximum covariance value is set. When the initial estimate is certain, the minimum covariance value is set, and the Kalman filter algorithm is used for iterative calculation.

7. The automatic time calibration system for power plant electric energy meters according to claim 6 is characterized in that: The specific method in which the correction error analysis unit performs iterative calculation using the Kalman filter algorithm is: For the prediction stage: Calculate the predicted value of the state according to the state transfer equation At the same time, calculate the prediction covariance matrix P k|k-1 =FP k-1|k-1 F T +Q, where Q is the process noise covariance matrix, which specifically reflects the intensity of the process noise; For the update phase: calculate the Kalman gain K k =P K|K-1 H T (HP K|K-1 H T +R) -1 , where R is the measurement noise covariance matrix, and then the state estimate is updated according to the measurement value Finally update the covariance matrix P k|k =(IK k H)P k|k-1 , where I represents the identity matrix, and (IK k H) is the Kalman gain K k The product of the measurement matrix H is corrected; For the clock compensation phase: the optimal estimate of the clock offset obtained in each iteration The clock is compensated by adjusting the clock time. The corresponding compensation amount is The hardware error impact factor and the software error impact are summed up to obtain a comprehensive error impact value. The comprehensive error impact value is used as a standard for time calibration and time calibration information is generated.

Images