KR102048429B1 - Power quality diagnosis system using frequency and demand correlation, and method thereof - Google Patents

Abstract

The present invention relates to a power quality diagnosis system which diagnoses quality of power supplied from a generator of a power plant and a method thereof. More specifically, the present invention relates to a power quality diagnosis system using a frequency and an output correlation coefficient of power and a method thereof. The power quality diagnosis system determines whether power is supplied by governor free (GF) operation or automatic generation control (AGC) operation by a frequency and an output correlation coefficient of the power supplied from a power plant and diagnoses and predicts quality of the power generated by one of the determined GF operation or AGC operation.

Description

Power quality diagnosis system using frequency and demand correlation, and method

The present invention relates to a power quality diagnosis system and method for diagnosing the quality of power supplied from a generator of a power plant. More specifically, the power is governed by a frequency and output correlation coefficient of power supplied from the power plant. Free: GF) determines whether it is supplied by the operation or Automatic Generation Control (AGC) operation, and the quality of the power produced by any one of the determined GF operation and AGC operation. The present invention relates to a power quality diagnosis system and method using frequency and output correlation coefficients of power for diagnosis and prediction.

 In general, the power industry has been privatized to improve the efficiency of power supply and to increase the convenience of consumers in the system that the government has monopolized. Before the privatization, the power industry had been involved in government regulation and regulation on the basis of supply and demand stability and publicity. After the privatization, the power industry is undergoing a big change, reorganizing into a competitive system that emphasizes efficiency and economy. Recently, the power industry has established a competitive power market environment in the midst of changes in the power system operation structure represented by the launch of power generation companies. Economical plant operation and power quality are therefore a very important issue in the competitive power market.

The government pays for the frequency adjustment operation to maintain the stability and reliability of the power system.

The frequency adjustment operation method includes Governor Free (GF) operation and Automatic Generation Control (AGC) operation.

In particular, when the power plant is divided into five subsidiaries from KEPCO, maintaining the rated frequency (60 Hz) by measuring and monitoring the speed regulation rate and output change rate according to the frequency change of the generator is one of the very important factors for the profit of the power plant. Is an important issue that can improve national competitiveness due to the improvement of electric quality.

For this purpose, the measurement of speed regulation rate and output rate of change is very important, but the conventional measurement of speed regulation rate and rate of change of output rate is dependent on manual labor, which frequently causes the problem of not meeting the frequency adjustment operation standard. It is not only possible to act as a system instability, but also to reduce the profits of power generation companies.

Due to this problem, Korean Patent No. 10-0668685 [Method for calculating the speed regulation rate and output change rate of a generator in a thermal power plant and a computer-readable recording medium recording a program for realizing the method] (hereinafter, "prior art" Discloses a technique for calculating the speed adjustment rate and the output change rate in real time.

However, since the prior art calculates only the speed adjustment rate and output change rate in real time, and cannot predict and inform the power quality, it is cumbersome from the manager side because the administrator must judge the quality of the power by measuring the measured speed adjustment rate and output change rate. There was a problem that it takes more time and money to lift.

Republic of Korea Patent No. 10-0668685 (January 15, 2007)

Accordingly, an object of the present invention is to determine whether the electric power is supplied by Governor Free (GF) operation by the frequency and output correlation coefficient of the electric power supplied from a power plant by the Automatic Generation Control (AGC) operation. The present invention provides a power quality diagnosis system and method using a frequency and an output correlation coefficient of power for determining whether to be supplied and for diagnosing and predicting the quality of power produced by one of the determined GF operation and AGC operation. .

Power quality diagnostic system using the frequency and output correlation coefficient of the power according to the present invention for achieving the above object: Sample data for storing a plurality of sample data for the frequency and output of power supplied from at least one power plant A storage unit including a DB; An alarm unit for notifying any one of normal, warning, and alarm for any one or more of the current state and the predicted state of the supplied electric power; And calculating a frequency correlation coefficient and an output correlation coefficient for each frequency and output by loading a predetermined number of sample data from the sample data DB, and comparing the frequency correlation coefficient and the output correlation coefficient to determine a governor free (GF) section. Identifying and diagnosing the quality by the frequency average value for the effective number of samples in the GF interval of the first number of continuous rising or falling in a certain number of sample data of the GF interval, and then the alarm unit according to the diagnosed quality It characterized in that it comprises a control unit for notifying the diagnosis result through.

The alarm unit notifies any one of normal, warning, and alarm for any one or more of a current state and a predicted state of power for each type of operation of the supplied power, and the controller controls the frequency correlation coefficient and the output correlation coefficient. Compare and identify the GF section and the automatic power control (AGC) section, diagnose the quality by the frequency average value corresponding to each of the identified GF section and the AGC section, and according to the quality and the predicted quality of the diagnosed power operation type It is characterized by controlling the alarm unit.

The alarm unit is characterized in that the audio processing unit for outputting the normal, warning and alarm by sound or voice.

The alarm unit is characterized in that the display unit for displaying any one or more of the normal, warning and alert text, graphics, still images and moving pictures.

The control unit may include: a sample extracting unit extracting and loading a predetermined number of sample data from the sample data DB; A correlation coefficient measuring unit calculating a frequency and an output correlation coefficient of the extracted sample data; A driving type determination unit comparing the frequency correlation coefficient with the output correlation coefficient to determine a GF section if the frequency correlation coefficient is large and a AGC section if the output correlation coefficient is large; An effective section number counting unit configured to calculate an effective section sample number of a GF section and an AGC section according to the determined driving type; The frequency average value is calculated based on the number of samples of the validity interval, and the quality average for each operation type is diagnosed according to the reference average value exceeded by comparing the calculated frequency average value with the alarm reference average value and the warning reference average value. A quality prediction unit; And an alarm processing unit for outputting a quality result by controlling the alarm unit according to the diagnosed result.

The control unit may further include a speed adjustment rate calculation unit that calculates a speed adjustment rate for sample data of the GF period when it is determined that the GF period.

If the number of valid section counts is determined to be a GF section, the number of valid section samples, which is the number of consecutive ups or downs of the first number of sample data within a predetermined number of sample data of the GF section, is counted and determined to be an AGC section. If the one of the output value and the output correlation coefficient within the predetermined number of sample data of the AGC interval is characterized in that the number of valid interval sample which is the number of sample data larger than the preset output reference value.

The quality predicting unit performs the quality diagnosis for each driving type in units of a first quality diagnosis cycle time, and when the count time exceeds the first quality diagnosis cycle time, the number of valid period samples of the AGC section counted is counted in all AGC sections. Calculate the AGC Value divided by the number and the GF Value obtained by dividing the number of effective interval samples of the GF interval by the total number of samples in the GF interval, and determining whether each of the AGC Value and the GF Value continuously increased or decreased during the second cycle time. Or if lowered, calculates a changed mean value for each of the AGC Value and GF Value for the first quality diagnostic period plus the second cycle time, and calculates the changed mean value and preset normal, warning and alarm thresholds. Forecast power quality compared to, and if yes or today for each of the AGC Value and GF Value, Comparing the change amount of each of the AGC Value and the GF Value, the change alarm reference value, the change warning reference value and the change normal reference value to predict the power quality for each AGC operation and GF operation according to whether the reference value is exceeded or less than do.

The cyste may further include a sample measuring unit connected to a generator of a power station and measuring power output from the generator to output sample data including the frequency and the output, wherein the control unit receives the sample data from the sample measuring unit. It further comprises a sample data acquisition unit for receiving the input and storing in the sample data DB.

Power quality diagnostic method using the power frequency and output correlation coefficient according to the present invention for achieving the above object: a sample extraction process by the control unit extracts and loads a predetermined number of sample data from the sample data DB through a sample extraction unit; ; A correlation coefficient measuring step of calculating, by the control unit, a frequency correlation coefficient and an output correlation coefficient of the extracted sample data through a correlation coefficient measuring unit; A driving type determination process of the control unit comparing the frequency correlation coefficient with the output correlation coefficient through a driving type determination unit to determine a GF section when the frequency correlation coefficient is large, and determining an AGC section when the output correlation coefficient is large; An effective section number counting step of the control unit counting the number of valid sample data according to the operation type determined by the valid section number counting unit and calculating the number of valid section samples in the GF section and the valid section samples in the AGC section; The control unit calculates a frequency average value based on the number of samples of the validity interval through the quality prediction unit, compares the calculated frequency average value, the alarm reference average value, and the warning reference average value to normal, warning and A quality diagnostic process for diagnosing any one of the alarms; And an alarm process in which the control unit outputs a quality result by controlling the alarm unit according to the predicted result through the alarm processing unit.

The correlation coefficient measurement process may include: calculating a covariance for each of a frequency and an output of the sample data; A standard deviation calculation step of calculating a sample standard deviation for each of the frequency and the output; And a correlation coefficient calculation step of calculating a frequency correlation coefficient and an output correlation coefficient for each of frequency and output by applying the covariance and the sample standard deviation.

The method may further include a speed adjustment rate calculation process of calculating a speed adjustment rate for the sample data of the GF period when the control unit determines that the GF period is determined in the driving type determination process.

In the validity section counting process, if the control unit determines that the GF section is a valid section sample by counting the valid section, which is the number of consecutive rising or falling first sample data within a certain number of sample data of the GF section, A governor-free validity period counting step of counting the number; And when the control unit determines that the AGC section is an AGC section, the automatic power control valid section which counts the number of valid section samples in which one of the output value and the output correlation coefficient is larger than a preset output reference value within a predetermined number of sample data of the AGC section. And a counting step.

The method may further include a quality prediction process for predicting the quality power for each of the AGC driving power and the GF driving power according to the AGC section and the GF section, wherein the quality prediction process includes a count time of the first quality diagnosis cycle time. Calculating a AGC value obtained by dividing the counted effective interval sample number of the AGC interval by the total number of samples in the AGC interval and a GF Value obtained by dividing the effective interval sample number of the GF interval by the total number of samples in the GF interval; It is determined whether each of the AGC Value and the GF Value continuously rises or falls during the second cycle time, and if the AGC Value and the GF Value have risen or fallen, the AGC Value and GF during the time of adding the second cycle time to the first quality diagnosis cycle. An AGC quality prediction step of calculating a changed average value for each Value and predicting the power quality by comparing the calculated changed average value with preset normal, warning and alarm thresholds; And if it is not rising or falling, comparing the AGC Value and the GF Value with the change amount of the AGC Value and the GF Value of yesterday and today, the change alarm reference value, the change warning reference value, and the change normal reference value, respectively; It characterized in that it comprises a GF quality prediction step of predicting the power quality for each of the AGC operation and GF operation by less than.

The method includes: a sample measuring process of connecting the control unit to a generator of a power plant through a sample measuring unit and measuring power output from the generator to output sample data including the frequency and the output; And acquiring a sample data in which the control unit stores the sample data in a sample data DB.

The method may further include a power management process of storing and managing power quality result information of the quality result output by the controller in a power management DB.

The present invention can confirm whether the power supplied from the current power plant is supplied by the GF operation or AGC operation, as well as diagnose and predict the power quality of the power supplied by the GF operation and the AGC operation. There is.

The present invention can provide the administrator with convenience because it can diagnose and predict the quality of the power supplied from the power plant in near real time, reduce the maintenance cost for quality control, and obtain the accurate sales amount according to the power quality. As you can, there are benefits that can improve your profits.

1 is a diagram illustrating a connection relationship between a power quality diagnosis system and a power plant using a frequency and an output correlation coefficient of power according to the present invention.
2 is a view showing a detailed configuration of a power quality diagnostic system using the power frequency and output correlation coefficient according to the present invention.
3 is a diagram illustrating power quality result information of each of a GF section and an AGC section according to an embodiment of the present invention.
4 is a flowchart illustrating a power quality diagnosis method using a frequency and an output correlation coefficient of power according to the present invention.
5 is a flowchart illustrating a method for setting an operation section in a power quality diagnosis method using a frequency and an output correlation coefficient of power according to the present invention.
6 is a flowchart illustrating a quality diagnosis method by GF operation among the quality prediction methods according to the present invention.
7 is a flowchart illustrating a quality diagnosis method by AGC operation of the quality prediction method according to the present invention.
8 is a flowchart illustrating a power quality prediction method for AGC operation among power quality diagnosis methods using a frequency and an output correlation coefficient of power according to the present invention.
9 is a flowchart illustrating a power quality prediction method for GF operation among power quality diagnosis methods using a frequency and an output correlation coefficient of power according to the present invention.

Hereinafter, a configuration and operation of a power quality diagnosis system using a frequency and an output correlation coefficient of power according to the present invention will be described in detail with reference to the accompanying drawings, and a power quality diagnosis method in the system will be described.

1 is a diagram illustrating a connection relationship between a power quality diagnosis system and a power plant using a frequency and an output correlation coefficient of power according to the present invention.

Referring to FIG. 1, the power plant 100 includes a generator 110 and a governor 120. The governor 120 receives a control command from the manager, that is, a speed adjustment rate, and the like, and outputs a control signal according to the speed adjustment rate of the manager to the generator 110 to the generator 110.

The generator 110 receives a control signal according to the speed adjustment rate from the governor 120 and generates power to reflect the speed adjustment rate and outputs it to the load.

The power quality diagnosis system 200 according to the present invention is connected to the output terminal of the generator 110 receives the power supplied to the load to measure the frequency and output of the power to give the sample data including the frequency and output (value) And continuously and continuously, after calculating the frequency correlation and the output correlation for the output frequency of the obtained sample data, the operation type is determined by the frequency correlation and the output correlation, and the power quality according to the operation type Diagnose and predict. The output means the magnitude of the load power. Detailed configuration and operation of the power quality diagnostic system 200 will be described in detail with reference to FIG.

2 is a diagram showing a detailed configuration of a power quality diagnosis system using a frequency and an output correlation coefficient of power according to the present invention, and FIG. 3 shows power quality result information of each of a GF section and an AGC section according to an embodiment of the present invention. The figure which shows. A description with reference to FIGS. 2 and 3 is as follows.

The power quality diagnostic system 200 according to the present invention includes a storage unit 210, an alarm unit 230, and a control unit 240, and may further include a sample measuring unit 220 according to an embodiment.

The storage unit 210 is a program area for storing a control program for controlling the operation of the power quality diagnostic system according to the present invention, a temporary area for temporarily storing data generated during the execution of the control program, occurs during the execution of the control program And a data area for semi-permanently storing the data and data necessary for driving the control program. In the data area, a sample data DB 211 storing a sample data string including a frequency and an output continuously measured at a predetermined period from power according to the present invention may be configured, and an AGC section as shown in FIG. 3 according to the present invention. And a power management DB 212 that stores quality prediction result information for the GF section. In addition, the sample data DB 211 may further include operation setting information set for each sample data. The operation setting information will indicate whether the corresponding sample data is sample data of the GF (driving) section or sample data of the AGC (driving) section. In addition, the data area has a reference value for determining each operation state, such as alarm, warning, normal according to the power operation method according to the present invention, that is, the alarm reference average value, warning reference average value, during the first cycle period + second cycle AGC Value Reference Value, Warning AGC Value Reference Value, Normal AGC Value Reference Value, Alarm GF Value Reference Value, Warning GF Value Reference Value, Normal GF Value Reference Value, Change Alarm Reference Value for Daily Change Valid Period Reference values, change normal reference values, etc. may be stored. The average value is calculated by AGC and GF {(number of samples valid for 2 days ago-number of samples valid for 2 days ago) + (number of samples valid for 1 day ago-number of samples valid for 1 day ago)} / 2.

The sample measuring unit 220 includes a power frequency measuring unit 221 for measuring the frequency of the power output from the generator 110 and a power output measuring unit 222 for measuring and outputting an output having a magnitude of the power. The sample data including the frequency and the output for the continuously generated at a predetermined period and outputs to the control unit 240.

The alarm unit 230 may notify any one of normal, warning, and alarm for any one or more of the current state and the predicted state of the GF operation of the power supplied according to the embodiment, or the current for each of the GF operation and the AGC operation. Notify any one of normal, warning, and alarm for any one or more of status and predicted status.

The alarm unit 230 may be an audio processor that outputs the normal, warning, and alarm in a sound or voice, or may be a display unit displaying the normal, warning, and alarm as one or more of text, graphics, still images, and video. In addition, it may be a communication means for notifying an external manager terminal (not shown) of normal, warning, and alarm for each operation type of current power.

The controller 240 includes a sample extractor 260, a correlation coefficient measurer 270, a driving type determiner 280, an effective section count counter 310, a quality predictor 320, and an alarm processor 330. In addition, according to the embodiment further comprises a sample data acquisition unit 250 and the speed adjustment rate calculation unit 290 to control the overall operation of the power quality diagnostic system according to the present invention.

Specifically, the sample data acquisition unit 250 obtains the sample data through the sample measuring unit 220 or obtains the sample data from the administrator through a communication means (not shown) and an input means (not shown) to obtain a sample data DB ( 211).

The sample extractor 260 extracts and loads a predetermined number of sample data from the sample data DB 211. For example, the predetermined number may be 120. That is, the sample extractor 260 extracts sample data from the sample data DB 211 in 120 units every time a quality diagnosis event occurs. The sample extractor 260 may extract a predetermined number of sample data regardless of the driving type according to the obtained order, or may extract a predetermined number of sample data for each specific driving type.

The correlation coefficient measurer 270 includes a covariance calculator 271, a standard deviation calculator 272, and a correlation coefficient calculator 273 to calculate the frequency and output correlation coefficients of the extracted sample data.

The covariance calculator 271 calculates the frequency covariance Cov _f (X, Y) by applying the frequency (x) and output (y) of the sample data to Equation 1 below, and outputs the sample data (x). ) and the load power (to the load Demand) (y) applied to the equation (1) to the output covariance (Cov _o (X, y), calculates a). The load power may be an output reference power as a target load power value.

The standard deviation calculator 272 applies the frequency (x) and output (y), the output (x) and the load power (y) of the sample data to the frequency covariance and the output covariance, respectively, Calculate the sample standard deviation for each of frequency and power, and power and load power.

The correlation coefficient calculator 273 calculates the frequency correlation coefficients P _{f_x, y} and the output correlation coefficients P _{o_x, y} by applying the calculated covariance and the sample standard deviation to Equation 3 below.

The driving type determination unit 280 compares the frequency correlation coefficient and the output correlation coefficient to determine whether the current sample data is a section (GF section) of power supplied by the GF operation or a section (AGC) of power supplied by the AGC operation. Section). That is, the driving type determination unit 280 determines whether the section of the current sample data is the sample data of the GF section or the sample data of the AGC section. The driving type determination unit 280 determines that the frequency correlation coefficient is a GF section, and when the output correlation coefficient is large, the driving type determination unit 280 determines a AGC section.

The speed adjustment rate calculator 290 calculates and outputs the speed adjustment rate of the current sample data in the GF section. The speed adjustment rate may be calculated by Equation 4 below.

Where P _rate is the generator rated output, Δf is the frequency change, and ΔP is the generator output change

The valid section number counting unit 310 counts and outputs the number of valid sample data according to the determined driving type.

The effective section number counting unit 310 counts the number of sample data rising or falling more than 4 samples in the GF section in the GF section, and output value or output correlation coefficient value in the AGC section in the AGC section. Count the number of out of sample data.

The quality estimator 320 may determine a value of the effective interval sample rate for each of the GF interval and the AGC interval that are changed during the frequency average value based on the number of effective interval samples, the first diagnosis cycle time, and the second cycle time. The cycle time average value and daily conversion amount are calculated, and the frequency average value, the alarm reference average value and the warning reference average value are compared, and the cycle time average value, daily change amount and normal reference value, warning reference value and alarm reference value, normal change amount reference value and alarm Diagnose and predict the quality of any one of normal, warning, and alarm for each operation type according to the reference value over which the calculated value is exceeded.

A detailed method for power quality diagnosis of the quality predicting unit 320 will be described in detail with reference to FIGS. 6 and 7.

A detailed method for power quality prediction of the quality predicting unit 320 will be described in detail with reference to FIGS. 4, 8, and 9.

The quality predicting unit 320 calculates AGC value and GF value in the AGC valid period sample number and the GF valid period sample number for each AGC operation and GF operation, for power quality prediction.

The AGC Value is the ratio of AGC valid interval samples divided by the number of AGC valid interval samples by the total number of samples in the AGC interval, and the GF Value is the ratio of GF valid interval samples divided by the number of GF valid interval samples divided by the number of samples in the entire GF interval.

 The quality predicting unit 320 generates power quality result information as shown in FIG. 3 including the calculated speed adjustment rate and diagnosis and predicted power quality information, and stores the same in the power management DB 212.

The alarm processor 330 controls the alarm 230 according to the predicted result and notifies the predicted quality result.

4 is a flowchart illustrating a power quality diagnosis method using a frequency and an output correlation coefficient of power according to the present invention, which is a flowchart showing a case including a sample measuring unit 220, and FIG. 8 is a frequency of power according to the present invention. And a power quality prediction method for AGC operation among power quality diagnostic methods using an output correlation coefficient, and FIG. 9 is a power for GF operation among power quality diagnostic methods using a frequency and an output correlation coefficient of power according to the present invention. A flowchart illustrating a quality prediction method. A description with reference to FIGS. 4, 8 and 9 is as follows.

First, the control unit 240 initializes the variables and settings of the system (S111).

The controller 240 continuously measures the frequency and output of the power output from the generator 110 through the sample measuring unit 220 (S113).

When the frequency and the output starts to be measured, the controller 240 generates sample data including the frequency and the output and stores the sample data in the sample data DB 211 of the storage 210 (S115).

When one piece of sample data is generated and stored, the controller 240 drives a timer to count the time (S117), and determines whether the first quality diagnosis period is based on the counted time (S125). If it is not the first quality diagnosis cycle, the controller 240 repeats the process after S117.

In addition, the control unit 240 monitors whether a quality diagnosis event occurs (S119), and when the quality diagnosis event occurs, each sample in units of a predetermined interval with respect to the power output from the generator 110 based on the stored sample data. After setting the operation method of the data (S121), a quality diagnosis is performed for each set operation method (S123). The quality diagnosis event may be configured to occur in units of intervals. The driving method setting will be described in detail with reference to FIG. 5, and the section quality diagnosis method will be described in detail with reference to FIGS. 6 and 7.

When the operation method setting and the section quality diagnosis start to be performed each time the quality diagnosis event occurs, the controller 240 checks whether the first quality diagnosis cycle time is exceeded by the counted time (S125). The first quality diagnostic cycle time may be 1 hour, 6 hours, 12 hours, 24 hours, and the like, and may be preferably 24 hours, that is, 1 day. It would be desirable. For example, the frequency of occurrence of the quality diagnostic event may be a time in minutes.

If the count time does not exceed the quality diagnostic cycle time, the control unit 240 repeats the above-described process of step S117, and the number of AGC valid interval samples (the ratio of the number of AGC non-standard samples to the defective samples) and the number of GF valid interval samples is repeated. (The ratio of the number of samples used in the rate adjustment rate analysis to the defective sample), and counting time exceeds the first quality diagnostic cycle time, calculates the AGC Value and GF Value (S127).

The AGC Value represents the AGC validity sample rate obtained by dividing the number of AGC validity interval samples by the total number of AGC interval samples (the number of AGC validity interval samples / the number of AGC validity samples).

In addition, GF Value represents the ratio of GF valid interval samples (speed adjustment rate sample ratio) obtained by dividing the number of GF valid interval samples by the total number of GF interval samples (the number of GF effective interval samples / the total number of GF interval samples).

When the AGC Value and the GF Value are calculated, the controller 240 outputs the state prediction and the prediction state for the AGC operation by the AGC Value (S129), and outputs the state prediction and the prediction state for the GF operation by the GF Value ( S131).

A state prediction and output process for AGC operation by AGC value will be described with reference to FIG. 8, and a state prediction and output process for GF operation by GF Value will be described with reference to FIG. 9.

Referring to FIG. 8, the controller 240 determines whether the AGC Value continuously rises or falls in the second cycle time (S141). The second cycle time may be set to a period equal to or greater than the first quality diagnosis attention time such as 2 days, 3 days, 4 days, 1 week, and the like.

If the AGC value continuously increases during the second cycle time, the average quality prediction process is performed (S143). If the AGC value is maintained without rising or falling continuously during the second cycle time, the daily variation amount quality prediction process is performed. S147).

The average quality prediction process, the control unit 240 calculates the average value of the daily AGC value changed during the time of adding the second cycle time to the first quality diagnostic cycle time (S141), the calculated average value and the preset alarm AGC Value The reference value, the warning AGC Value reference value, the normal AGC Value reference value is compared and the current state is predicted and displayed in the corresponding quality state among the alarm, warning, and normal (S145).

On the other hand, in the change quality prediction process, the control unit 240 calculates the change amount AGC Value subtracted from the today's AGC Value by the daily change amount subtracted yesterday and today's AGC Value (S147).

When the change amount AGC value is obtained, the control unit 240 compares the change amount AGC value with a pre-stored alarm change amount reference value, a warning change amount reference value, and a normal change amount reference value, respectively, and predicts and displays a quality state corresponding to the corresponding state (S149).

Referring to FIG. 9, the controller 240 determines whether the GF Value continuously rises or falls in the second cycle time (S151). The second cycle time may be set to a period equal to or greater than the first quality diagnosis attention time such as 2 days, 3 days, 4 days, 1 week, and the like.

If the GF value is continuously increased during the second cycle time, the average quality prediction process is performed (S153), and if the GF value is maintained without rising or falling continuously during the second cycle time, the daily variation amount quality prediction process is performed ( S157).

The average quality prediction process, the control unit 240 calculates the average value of the daily GF Value changed during the time of adding the second cycle time to the first quality diagnostic cycle time (S151), the calculated average value and the preset alarm GF Value Comparing the reference value, warning GF Value reference value, normal GF Value reference value and predicts and displays the current state in the corresponding quality state of alarm, warning, normal (S155).

On the other hand, in the change quality prediction process, the control unit 240 calculates the change amount GF Value subtracted by the daily change amount by subtracting yesterday and today's GF Value from today's GF Value (S157).

When the change amount GF Value is obtained, the control unit 240 compares the change amount GF Value with a pre-stored alarm change amount reference value, a warning change amount reference value, and a normal change amount reference value, respectively, and predicts and displays a quality state corresponding to the corresponding state (S159).

5 is a flowchart illustrating a method for setting an operation section in a power quality diagnosis method using a frequency and an output correlation coefficient of power according to the present invention.

Referring to FIG. 5, the controller 240 extracts a predetermined number of sample data stored in the sample data DB 211 (S211). The number of sample data may be 120.

When the sample data is extracted, the controller 240 calculates a frequency covariance and an output covariance for each of the frequencies and outputs of the extracted 120 units of sample data, and calculates a sample standard deviation for each of the frequencies and outputs x and y ( S213, S215).

When the covariance and the sample standard deviation are calculated, the control unit 240 calculates the frequency correlation coefficient and the output correlation coefficient for each of the frequency and the output to which the covariance and the sample standard deviation are applied (S217), and then determine whether the frequency correlation coefficient is greater than the output correlation coefficient. Determine the type of operation by determining (S219).

The controller 240 sets the AGC section when the output correlation coefficient is greater than the frequency correlation coefficient (S221), and sets the GF section when the frequency correlation coefficient is larger than the output correlation coefficient (S225).

When the section according to the driving type according to the driving type starts to be set, the controller 240 determines whether the driving section according to the driving type is set for all the extracted samples (S227), and the first quality diagnosis period. The above procedure is repeated for all samples in time.

6 is a flowchart illustrating a quality diagnosis method by GF operation among the quality prediction methods according to the present invention.

Referring to FIG. 6, as shown in FIG. 5, when the section setting according to the driving type is completed in the section unit including the predetermined sample data, the control unit 240 receives a predetermined number of sample data set as the GF section in the sample data DB 211. Extracted from (S310).

When a predetermined number of sample data set as the GF section is extracted, the control unit 240 considers a section in which four samples are continuously rising or falling as a valid section, and counts the number of valid section samples which is the number of sample data of the valid section (S311).

When the valid period sample number for the extracted sample number is counted, the controller 240 calculates and analyzes the speed adjustment rate for the counted valid period sample data and stores the calculated speed adjustment rate and analysis information (S313).

After calculating the speed adjustment rate and storing the analysis information, the controller 240 calculates a frequency average value for the entire GF section (S314).

When the frequency average value for the entire GF section is calculated, the controller 240 checks whether the frequency average value exceeds the alarm reference average value (S315) or exceeds the warning reference average value (S317).

When the frequency average value exceeds the warning reference average value, the control unit 240 alerts the current state to the alarm state (S319). If the average value exceeds the warning reference average value, the controller 240 indicates the current quality state to the warning state (S321). After determining the current quality state to the normal state (S323), the power quality diagnosis result is notified through the alarm unit 230 (S325).

7 is a flowchart illustrating a quality diagnosis method by AGC operation of the quality prediction method according to the present invention.

Referring to FIG. 7, when the section setting according to the driving type is completed in the section unit including the schedule sample data as shown in FIG. 5, the control unit 240 receives a predetermined number of sample data set as the AGC section in the sample data DB 211. Extract from (S411).

When a predetermined number of sample data set in the AGC section is extracted, the controller 240 determines whether any one of the output correlation and the output is out of the preset output reference range (S413). The output reference range may be defined to have a K range value based on the output reference value or may be defined as a W output range.

The control unit 240 counts the number of valid period samples, which is the number of sample data outside the output reference range (S415).

When the number of valid interval samples is counted, the controller 240 calculates a frequency average value for the entire AGC interval (S416).

The controller 240 notifies the power quality diagnosis result of the sample data of the AGC section by performing S417 to S429 corresponding to each of S315 to S325 described above with reference to FIG. 6.

On the other hand, the present invention is not limited to the above-described typical preferred embodiment, it can be carried out in various ways without departing from the gist of the present invention various modifications, alterations, substitutions or additions are common knowledge in the art Anyone who has this can easily understand it. If such improvement, change, substitution or addition is carried out within the scope of the appended claims, the technical spirit should also be regarded as belonging to the present invention.

100: power plant 110: generator
120: governor 200: power quality diagnostic system
210: storage unit 211: sample data DB
212: power management DB 220: sample measuring unit
221: power frequency measurement unit 222: power output measurement unit
230: alarm unit 240: control unit
250: sample data acquisition unit 260: sample extraction unit
270: correlation coefficient measuring unit 271: covariance calculation unit
272: standard deviation calculation unit 273: correlation coefficient calculation unit
280: operation type determination unit 290: speed adjustment rate calculation unit
310: Valid section count counter 320: Quality predictor
330: alarm processing unit

Claims (16)

A storage unit including a sample data DB for storing a plurality of sample data about a frequency and an output of power supplied from at least one power plant;
An alarm unit for notifying any one of normal, warning, and alarm for any one or more of a current state and a predicted state of power for each type of operation of the supplied power; And
Load a predetermined number of sample data from the sample data DB to calculate the frequency correlation coefficient and output correlation coefficient for each frequency and output, and compare the frequency correlation coefficient and the output correlation coefficient to the GF interval and automatic power control (AGC) And a controller for identifying the section, diagnosing the quality by the frequency average value corresponding to each of the identified GF section and the AGC section, and notifying the diagnosis result through the alarm unit according to the quality and the predicted quality of the diagnosed power operation type. But
The control unit,
A sample extractor for extracting and loading a predetermined number of sample data from the sample data DB;
A correlation coefficient measuring unit for calculating a frequency correlation coefficient and an output correlation coefficient of the extracted sample data;
A driving type determination unit comparing the frequency correlation coefficient with the output correlation coefficient to determine a GF section if the frequency correlation coefficient is large and a AGC section if the output correlation coefficient is large;
An effective section number counting unit configured to calculate an effective section sample number of a GF section and an AGC section according to the determined driving type;
The frequency average value is calculated based on the number of samples of the validity interval, and the quality average for each operation type is diagnosed according to the reference average value exceeded by comparing the calculated frequency average value with the alarm reference average value and the warning reference average value. A quality prediction unit; And
Power quality diagnostic system using the frequency and output correlation coefficient of the power, characterized in that it comprises an alarm processing unit for outputting a quality result by controlling the alarm in accordance with the diagnosis result.
delete The method of claim 1,
The alarm unit,
Power quality diagnostic system using the frequency and output correlation coefficient of the power, characterized in that the audio processing unit for outputting the normal, warning and alarm by sound or voice.
The method of claim 1,
The alarm unit,
And a display unit for displaying the normal, warning, and alarm in any one or more of text, graphics, still images, and moving images.
delete The method of claim 1,
The control unit,
The power quality diagnosis system using the frequency and output correlation coefficient of the power, characterized in that it further comprises a speed adjustment rate calculation unit for calculating the speed adjustment rate for the sample data of the GF period when determined as the GF period.
The method of claim 1,
The effective section number counting unit,
If it is determined that the GF interval, the number of valid interval samples, which is the number of consecutive rising or falling first sample data in a certain number of sample data of the GF interval is counted,
If it is determined that the AGC interval, the frequency and output of power, characterized in that counts the number of valid interval samples, which is the number of sample data larger than the preset output reference value of any one of the output value and the output correlation coefficient within a certain number of sample data of the AGC interval Power Quality Diagnosis System Using Correlation Coefficient.
The method of claim 7, wherein
The quality prediction unit,
The quality diagnosis for each operation type is performed in units of a first quality diagnosis cycle time,
When the count time exceeds the first quality diagnostic cycle time, the AGC Value obtained by dividing the counted number of valid period samples of the AGC interval by the total number of samples in the AGC interval and the GF Value obtained by dividing the number of effective interval samples of the GF interval by the total number of samples in the GF interval Calculate,
It is determined whether each of the AGC Value and the GF Value continuously rises or falls during the second cycle time, and if the AGC Value and GF Value have risen or fallen, the AGC Value and GF during the time of adding the second cycle time to the first quality diagnosis cycle. Calculate the changed average value for each of the Values, predict the power quality by comparing the calculated changed average value with preset normal, warning and alarm thresholds,
If it is not rising or falling, the change amount of the AGC Value and the GF Value of yesterday and today, the change alarm reference value, the change warning reference value, and the change normal reference value for the AGC Value and GF Value, respectively, are compared or exceeded. Power quality diagnosis system using the frequency and output correlation coefficient of the power, characterized in that for predicting the power quality for each of the AGC operation and GF operation by whether or not.
The method of claim 1,
It further comprises a sample measuring unit connected to the generator of the power station for measuring the power output from the generator to output the sample data including the frequency and output,
The control unit,
And a sample data acquisition unit for receiving sample data from the sample measuring unit and storing the sample data in the sample data DB.
A sample extraction process by the controller extracting and loading a predetermined number of sample data from the sample data DB through the sample extractor;
A correlation coefficient measuring step of calculating, by the control unit, a frequency correlation coefficient and an output correlation coefficient of the extracted sample data through a correlation coefficient measuring unit;
The control unit compares the frequency correlation coefficient and the output correlation coefficient through a driving type determination unit, and if the frequency correlation coefficient is large, determines that the driving type is GF section having GF, and if the output correlation coefficient is large, determining that the driving type is AGC section having AGC. Driving type judgment process;
An effective section number counting process of the control unit counting the number of valid sample data according to the operation type determined by the valid section number counting unit and calculating the number of valid section samples in the GF section and the valid section samples in the AGC section;
The control unit calculates a frequency average value based on the number of samples of the validity interval through the quality prediction unit, compares the calculated frequency average value, the alarm reference average value, and the warning reference average value to normal, warning and A quality diagnostic process for diagnosing any one of the alarms; And
And an alarm process in which the control unit controls the alarm unit according to the predicted result and outputs a quality result through the alarm processing unit.
The method of claim 10,
The correlation coefficient measurement process,
Calculating a covariance for each of frequency and output of the sample data;
A standard deviation calculation step of calculating a sample standard deviation for each of the frequency and the output; And
And a correlation coefficient calculation step of calculating a frequency correlation coefficient and an output correlation coefficient for each of a frequency and an output by applying the covariance and the sample standard deviation.
The method of claim 10,
The power quality using the frequency and output correlation coefficient of the power, characterized in that further comprising the step of calculating the speed adjustment rate for the control step for the sample data of the GF section when the control unit is determined in the operation type determination process Diagnostic method.
The method of claim 10,
The valid section number counting process,
If the control unit determines that the GF section is a governor-free effective counting the valid section number by counting the valid section, which is the number of the first number of consecutive rising or falling sample data within a certain number of sample data of the GF section An interval counting step; And
When the control unit determines that the AGC section is an AGC section, the automatic power control valid section number counting the number of valid section samples, which is the number of sample data which is larger than a preset output reference value, among the predetermined number of sample data of the AGC section. Power quality diagnostic method using the frequency and the output correlation coefficient of the power, characterized in that it comprises a count step.
The method of claim 10,
Further comprising a quality prediction process for predicting the quality power for each of the AGC driving power and the GF driving power according to the AGC section and the GF section,
The quality prediction process,
If the count time exceeds the preset first quality diagnostic cycle time, the AGC Value obtained by dividing the counted number of valid interval samples of the AGC interval by the total number of samples in the AGC interval and the number of effective interval samples of the GF interval divided by the total number of samples in the GF interval A ratio calculation step of calculating a value;
It is determined whether each of the AGC Value and the GF Value continuously rises or falls during the second cycle time, and if the AGC Value and the GF Value have risen or fallen, the AGC Value and GF during the time of adding the second cycle time to the first quality diagnosis cycle. An AGC quality prediction step of calculating a changed average value for each Value and predicting the power quality by comparing the calculated changed average value with preset normal, warning and alarm thresholds; And
If it is not rising or falling, the change amount and change alarm reference value, change warning reference value and change normal reference value of the AGC Value and GF Value of yesterday and today, respectively, for the AGC Value and GF Value, respectively, are compared or exceeded. And a GF quality prediction step of predicting power quality for each of the AGC operation and the GF operation according to whether or not the power quality is diagnosed.
The method of claim 10,
A sample measurement process of the control unit being connected to a generator of a power plant through a sample measuring unit and measuring power output from the generator to output sample data including the frequency and the output; And
And a step of acquiring sample data in which the control unit stores the sample data in a sample data DB.
The method of claim 10,
And a power management process of storing and managing power quality result information of the quality result output by the control unit in a power management DB.

Images