KR20200039970A - System and method for transmission of power time series data - Google Patents

Abstract

The present invention relates to a transmission processing system of power time series data and to a method thereof. The system comprises: at least one Internet of Things (IoT) sensor module unit transmitting whether to analyze and transmit measurement data, and then transmitting the measurement data in accordance with the determination result; a power big data analysis unit receiving and analyzing power data from the at least one IoT sensor and generating time series statistics data for each detection target of a corresponding power generation facility; a client requesting data by connecting to the big data analysis unit and receiving and monitoring result data corresponding thereto; and a communication network connecting the IoT sensor, the power big data analysis unit, and the client by communication.

Description

System and method for transmission of power time series data {SYSTEM AND METHOD FOR TRANSMISSION OF POWER TIME SERIES DATA}

The present invention relates to a system and method for transmitting and processing power time series data, and more specifically, increases network load due to massive power data transmitted from at least one IoT (Internet of Things) sensor installed in a power generation facility. The present invention relates to a system and method for transmitting and processing power time series data, which can improve a problem.

In general, the industrial Internet (ie, the term used by General Electric, which is used in front of industrial sites such as manufacturing plants, medical, railways, and electric power) produces data at twice as fast as other big data fields. At the same time, since the generation cycle has a short characteristic, various studies have been conducted on technologies that can improve the efficiency of power big data processing by collecting and analyzing power big data derived from power generation facilities through high-speed and efficient techniques.

In the case of power energy data, which is an important field of the industrial Internet, the network load phenomenon (ie, overload) occurs or data storage occurs when data is transmitted to the outside as it is because the amount of high-precision time series data produced in real time is vast. It has the disadvantage of low efficiency.

Here, the time series data is data expressed as a function of time for a certain period of time, and is used for predicting future results using data generated according to the flow of time in the past.

Therefore, there is a need for an efficient transmission technology that determines whether time series data generated in the power industry environment is valid data or invalid data in advance, and then transmits only valid data. However, there have been no studies on the data transmission method of this method.

Background art of the present invention is disclosed in Republic of Korea Patent Registration No. 10-1318983 (October 10, 2013, uninterruptible power quality measurement device to notify whether the electricity meter is abnormal).

According to an aspect of the present invention, the present invention was created to solve the above problems, and the network load increases due to massive power data transmitted from at least one Internet of Things (IoT) sensor installed in a power generation facility. An object of the present invention is to provide a system and method for transmitting and processing power time series data.

In addition, according to an aspect of the present invention, the present invention specifies a condition in which power data is low need to be transmitted from at least one IoT sensor installed in a power generation facility, and then, when the condition is satisfied, the network data load is not transmitted. It is to provide a system and method for transmitting and processing power time-series data that can reduce the.

At least one or more of the transmission processing system for power time series data according to an aspect of the present invention measures a state of a power generation facility, analyzes this measurement data, determines whether to transmit it, and then transmits the measurement data according to the determination result. IoT (Internet of Things) sensor module unit; A power big data analysis unit that receives and analyzes power data from the at least one IoT sensor and generates time series statistical data for each detection target of the power generation facility; A client accessing the power big data analysis unit to request data, and receiving and monitoring result data corresponding thereto; And a communication network connecting the IoT sensor, the power big data analysis unit, and the client by communication.

In the present invention, the IoT sensor module, at least one of a physical sensor and a physical sensor, a physical sensor, an environmental sensor, a quality sensor, a fire sensor, and a physical and chemical sensor for detecting a state related to the operation of the power generation facility It is characterized by including one or more.

In the present invention, the IoT sensor module, the initial value and the threshold range according to the characteristics of the detection target is preset and stored, and analyzes the collected power data using the initial value and the threshold range to transmit this power data It is characterized by determining whether or not.

In the present invention, the initial value means a normal measurement value measured at a corresponding sensing target of the power generation facility, and the threshold range means a variation range that can be determined to have not changed from the initial value Is done.

In the present invention, the IoT sensor module transmits the power data when it is determined that the transmission of the power data measured by the power generation facility is meaningful, and transmits the power data when it is determined that the transmission of the power data is meaningless. However, when it is determined that the transmission of the power data is meaningful, it means that the power data is outside a predetermined threshold range, and when it is determined that the transmission of the power data is meaningless, the power data is preset It is characterized in that it means a case that does not deviate from the handover range.

In the present invention, the power big data analysis unit stores a predetermined initial value and a threshold range of the IoT sensor module, and searches whether power data is transmitted from the IoT sensor module every predetermined period (T), When power data is not transmitted, it is determined that power data of the corresponding IoT sensor module is included in a threshold range of the initial value, and power data of the corresponding IoT sensor module is determined as an initial value.

In the present invention, the power big data analysis unit uses the received power data as an input value, but uses the preset machine learning algorithm (Machine learning Algorithm) for short-term power prediction. Learning, analyzing the power data received from the IoT sensor module using the learned prediction pattern, predicting and diagnosing short-term power, and when the prediction result comes out, it is implemented to transmit prediction data to the client do.

In the present invention, the IoT sensor module, a communication module for transmitting and receiving data with the power big data analysis unit through a communication network; A memory for storing an initial value and a threshold range corresponding to the characteristics of the sensing target; And a control module configured to control sensing and transmission of sensed power data according to a designated method.

In the present invention, the control module, the sensing module for measuring the state of the preset detection target every period (T); A comparison module that is driven when power data is measured by the sensing module and compares whether the measured power data is included within a threshold range of an initial value stored in the memory; And when driving the comparison module, if the measured power data is included within a threshold range of the initial value, it is determined that the corresponding power data is not transmitted, and if the measured power data is outside the threshold range of the initial value, the corresponding power Characterized in that it comprises; a determination module for determining that the data should be transmitted.

In the present invention, the control module, when the power data measured by the sensing module for a predetermined number of times out of the threshold range from the initial value, and also when the deviation of the power data is a value within a predetermined threshold range And, it characterized in that it further comprises a; reset module for resetting the preset initial value to the average value of the power data.

In the present invention, the initial value of the IoT sensor module reset by the reset module is transmitted to a power big data analysis unit, and the power big data analysis unit resets the received reset initial value to the initial value of the corresponding IoT sensor module. It is characterized by resetting to.

In the present invention, the power big data analysis unit, a communication interface unit for receiving the power data of the IoT sensor module; Identification information of IoT sensor module, information about each power generation facility, initial value and threshold range of IoT sensor module, power data received from IoT sensor module, power data updated by power data update unit, preset analysis algorithm, machine A database server storing at least one of a running algorithm and a pattern detection algorithm; A power data update unit that replaces or updates a preset initial value or a reset initial value when power data is not transmitted from the IoT sensor module; A statistical data generation unit that generates time series statistical data by utilizing the power data received from the IoT sensor module and the power data updated by the power data update unit; A data analysis unit for analyzing time-series statistical data for a predetermined period using an analysis algorithm selected or registered by the client; A dashboard management unit managing a dashboard that outputs time series statistical data or analysis data generated according to the client's request; And a learning unit that uses power data as an input value and learns a power event prediction pattern using a preset machine learning algorithm aimed at power prediction.

In the present invention, the power data update unit, a search module for searching whether there is an inactive IoT sensor module that does not transmit power data for a specified period (T) of the IoT sensor module; An extraction module extracting an initial value or a reset initial value of the inactive IoT sensor module detected from the database server when the inactive IoT sensor module is detected by the search module; And a replacement module that replaces the initial value or the reset initial value extracted by the extraction module with power data of the corresponding inactive IoT sensor module.

In the present invention, the predictive analysis unit, by using a predetermined pattern detection algorithm, analyzes time series statistical data for a predetermined period of time to detect a pattern of each power generation facility pattern detection module; A similarity detection module that detects the similarity between the pattern detected by the pattern detection module and the event data learned by the learning unit; And a prediction module that predicts that a corresponding event will occur when the similarity is greater than or equal to a threshold by the similarity detection module.

A method for transmitting power time series data according to another aspect of the present invention includes: setting an initial value and a threshold range of a detection target in advance in an IoT sensor module; Measuring, by the IoT sensor module, power data of a preset sensing target every designated period (T); Comparing, by the IoT sensor module, whether the measured power data is outside a threshold range of the initial value; Determining that it is necessary to transmit the power data to the power big data analysis unit when the measured power data out of the threshold range of the initial value in the comparing step of the IoT sensor module; And determining that the corresponding power data is not to be transmitted to the power big data analysis unit when the measured power data is included within a threshold range of the initial value.

According to an aspect of the present invention, the present invention enables to improve the problem of an increase in network load due to massive power data transmitted from at least one Internet of Things (IoT) sensor installed in a power generation facility.

In addition, according to an aspect of the present invention, the present invention specifies a condition in which power data is low need to be transmitted from at least one IoT sensor installed in a power generation facility, and then, when the condition is satisfied, the network data load is not transmitted. Can be reduced.

1 is an exemplary view showing a schematic configuration of a transmission processing system for power time series data according to an embodiment of the present invention.
FIG. 2 is an exemplary view shown in FIG. 1 for explaining an operation method of a power time series data transmission processing system.
3 is an exemplary view showing a schematic configuration of an IoT sensor in FIG. 1.
4 is an exemplary view for explaining a schematic operation method of the IoT sensor in FIG. 3.
5 is a flowchart for schematically explaining the operation of the IoT sensor in FIG. 3.
FIG. 6 is an exemplary view showing a schematic configuration of a power big data analysis unit in FIG. 1.
7 is an exemplary view showing a more specific configuration of the power data update unit in FIG. 6.
8 is an exemplary view showing a more specific configuration of the prediction analysis unit in FIG. 6.

Hereinafter, an embodiment of a transmission processing system and method for power time series data according to the present invention will be described with reference to the accompanying drawings.

In this process, the thickness of the lines or the size of components shown in the drawings may be exaggerated for clarity and convenience. In addition, terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to a user's or operator's intention or practice. Therefore, the definition of these terms should be made based on the contents throughout the present specification.

The present invention compares the data measured by the IoT sensor with a preset initial value, and when it is within an initial value and a threshold range, the corresponding data is not transmitted to the power big data analysis unit, but the power big data analysis unit performs data for a specified period (T). When the IoT sensor that does not transmit is detected, the initial value of the IoT sensor is replaced with the data value to be transmitted, thereby minimizing the network load in data transmission to improve the efficiency of the big data transmission and management system.

Here, the data measured by the IoT sensor simultaneously senses not only power parameter data such as voltage, current, and frequency generated by the power generation module (or at least one power generation facility), but also environmental data including parameters such as temperature, humidity, and illuminance. Refers to the integrated time series data. Accordingly, `` power data (or power big data) '' described in the present embodiment means the integrated time series data (i.e., power parameter data + environment data), and for convenience of description, simply `` power data (or power big data) ''. It can be described collectively as'.

In addition, according to the present invention, the IoT sensor resets the initial value to the average value of the power data when the power data measured during the preset number of times is out of the threshold range of the initial value and the difference values of the power data are all within the threshold range. Prevent network overload.

Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

1 is an exemplary view showing a schematic configuration of a transmission processing system for power time series data according to an embodiment of the present invention, and FIG. 2 is a diagram illustrating an operation method of the transmission processing system for power time series data in FIG. 1. This is an example diagram shown for this purpose.

As shown in FIG. 1, the power time series data transmission processing system 1 according to the present embodiment transmits and processes power data collected from at least one power generation facility 4 to effectively operate and manage the power generation facility. At the same time, it reduces costs.

In addition, the transmission processing system for power time-series data according to the present embodiment predicts an event in advance, and compares the power data measured by the power IoT sensor 5 with an initial value, and when there is no significant change, the IoT sensor 5 By not transmitting the power data measured by) to the power big data analysis unit 3, network load (or overload) due to data transmission and reception is effectively reduced.

In addition, the transmission processing system for power time series data according to the present embodiment measures the state of at least one power generation facility 4, 4-1, 4-N generating power, as shown in FIGS. 1 and 2. It includes at least one IoT sensor (5). It also includes a power big data analysis unit 3 for analyzing power data (or power big data) transmitted from the at least one IoT sensor 5. It also includes a client 9 that connects to the power big data analysis unit 3, requests data, and receives and monitors result data corresponding to the requested data. It also includes a communication network 10 that provides a movement path for providing data measured from the IoT sensor 5 to the power big data analysis unit 3 and the client 9.

At this time, as shown in FIG. 2, the platform for providing and processing the data measured from the IoT sensor 5 to the power big data analysis unit 3 and the client 9 is a power platform as a service (PaaS). ) Use the platform. Here, the PaaS means a platform as a service, and means a system that provides hardware and software necessary for designing, developing, and distributing computer applications (applications) over the Internet.

For reference, the power generation facility 4 means a facility that generates power, such as a power plant, a substation, and a micro grid.

The IoT sensors 5 are installed in the power generation facility 4 to measure a preset object.

For example, the IoT sensors 5 may include a location sensor, a physical sensor, an environmental sensor, a quality sensor, and a fire sensor, and also various physical / for detecting a state related to the operation of the power generation facility. It can consist of chemical sensors.

In addition, in each of the IoT sensors 5, an initial value and a threshold range according to the characteristics of the sensing target are preset and stored, and the collected power data is analyzed using the initial value and the threshold range to determine whether to transmit power data. Judge.

Therefore, in the present embodiment, the IoT sensors 5 are not only meant to be sensing elements, but should be interpreted as sensor modules including control means for analyzing and transmitting whether to sense data measured through the sensing elements.

At this time, the initial value is defined as a normal measurement value (for example, an average value) measured at a corresponding sensing target of the power generation facility, and the threshold range is defined as a variation range that can be determined to have not changed from the initial value.

Therefore, the IoT sensor 5 measures a preset detection target of the power generation facility 4, but only when it is determined that it is meaningful after determining whether transmission of the measured power data is meaningful (for example, a measured value (i.e. , Measured power data) is out of the threshold range), by transmitting the power data to the power big data analysis unit 3, to prevent the transmission of meaningless power data to reduce the network load (overload) .

The power big data analysis unit 3 includes hardware and software such as an operating system, middleware, and processor, which form a basis on which an application program can be executed.

In addition, the power big data analysis unit 3 receives the power data from the IoT sensors 5, and stores the received power data in a local DB (not shown).

In addition, the power big data analysis unit 3 stores the predetermined initial values and threshold ranges of the IoT sensors 5, and whether power data is transmitted from the IoT sensors 5 every predetermined period T If the power data is not received from a specific IoT sensor, it is determined that the power data of the IoT sensor (i.e., the IoT sensor that does not receive the power data) is included in the threshold range of the initial value. The power data is determined as the initial value.

That is, in the present embodiment, it is determined whether or not transmission of the measured power data is performed by using the initial value and the threshold range set in each IoT sensor 5, and the power big data analysis unit 3 if the specific IoT sensor 5 When data reception from is omitted, it is possible to significantly reduce the network load by replacing the power data with a preset initial value.

In addition, the power big data analysis unit 3 generates time series statistical data for each detection target of the power generation facility by utilizing the power data stored for a predetermined period.

In addition, the power big data analysis unit 3 is pre-set and stored analysis algorithms for detecting valuable data by analyzing generated time series statistical data, and any one of the preset analysis algorithms from the client 9 When selected, time series statistical data generated using the selected analysis algorithm is analyzed.

In addition, the power big data analysis unit 3 uses the received power data as an input value, but uses the preset machine learning algorithm aimed at short-term power prediction to learn the input power data. You can.

In addition, the power big data analysis unit 3 may predict and diagnose short-term power by analyzing power data received from the IoT sensors 5 using a previously learned prediction pattern.

In addition, when the result of the predictive analysis comes out, the power big data analysis unit 3 transmits the predicted data to the client 9, so that a quick response to the prediction is made.

The client 9 supports the connection of the communication network 10, and provides factors according to the format of the power big data analysis unit 3, so that the power big data analysis unit 3 requests a service provided It is a digital terminal. The client 9 connects to the power big data analysis unit 3 to request data from the power big data analysis unit 3, and monitors calculation processing data received from the power big data analysis unit 3 (Not shown).

In addition, a web interlocking program (web browser, Netscape, etc.) is installed in the client 9 to access the power big data analysis unit 3 through the interlocking program.

At this time, the client 9 is provided with a wired communication module or a wireless communication module according to the type of the communication network 10, in detail, it can be applied to a desktop computer, notebook, PDA (Personal Digital Assistant), smart phone, tablet PC, etc. You can.

FIG. 3 is an exemplary view showing a schematic configuration of an IoT sensor (ie, a sensor module including control means for determining whether to analyze and transmit sensing data measured through a sensing element) in FIG. 1, FIG. 4 is an exemplary diagram for describing a schematic operation method of the IoT sensor in FIG. 3.

As illustrated in FIG. 3, the IoT sensor 5 includes a control module 51, a communication module 53, and a memory 55. In addition, the control module 51 includes a sensing module 511, a comparison module 513, a determination module 515 and a reset module 517.

The communication module 53 connects to the communication network 10 to transmit and receive data with the power big data analysis unit 3.

The memory 55 stores preset initial values and threshold ranges in consideration of characteristics of a sensing target measured by the IoT sensor 5.

The control module 51 is a microprocessor of the IoT sensor 5.

The sensing module 511 measures the state of a preset detection target according to a pre-designed period (T). At this time, the detection target includes temperature, environment, fire, speed, distance, and moving range.

The comparison module 513 is driven when power data is measured by the sensing module 511 and compares whether the measured power data is included within a threshold range of an initial value stored in the memory 55.

When the comparison module 513 is driven, the determination module 515 determines that the measured power data is not transmitted to the power big data analysis unit 3 if the measured power data is included within a threshold range of the initial value. do. In addition, the determination module 515 determines that the power data must be transmitted to the power big data analysis unit 3 when the measured power data is outside the threshold range of the initial value when the comparison module 513 is driven. do.

At this time, if it is determined that the corresponding power data should be transmitted by the determination module 515, the communication module 53 transmits the corresponding power data to the power big data analysis unit 3.

For example, referring to FIG. 4, when it is assumed that the detection target of the specific IoT sensor 5 is 'temperature', the initial value is set to 19 degrees, and the threshold range is set to 0 (①), if the IoT sensor 5 is When the measured power data (temperature) is 19 degrees (②), it is determined that the corresponding power data is not to be transmitted to the power big data analysis unit 3 and is not transmitted (③). However, if the measured power data (temperature) is 23 degrees, it is determined that the corresponding power data should be transmitted to the power big data analysis unit 3 (③) and transmitted through the communication module 53 (④). . Accordingly, the power big data analysis unit 3 checks the power data transmitted from the IoT sensor 5 again and performs pre-processing processing to store the result in the server (①, ⑤ to ⑦).

The reset module 517 when the power data measured by the sensing module 511 for a predetermined number of times is out of the threshold range from the initial value, and also if the deviation of the power data is within a predetermined threshold range , Reset the preset initial value to the average value of the power data.

Accordingly, when the comparison module 513 is operated later, the reset initial value is utilized.

In addition, the initial value reset by the reset module 517 is transmitted to the power big data analysis unit 3, and the power big data analysis unit 3 receives the reset initial value received by the corresponding IoT sensor 5 Reset to the initial value of. That is, the IoT sensor 5 according to the present embodiment resets the initial value to further increase the network load when the value of the power data measured by various causes such as the surrounding environment is continuously measured while having a large difference from the initial value. It can be effectively prevented.

5 is a flowchart for schematically explaining the operation of the IoT sensor in FIG. 3.

Referring to FIG. 5, each IoT sensor 5 presets an initial value and a threshold range of a sensing target (S110), and measures (sensing) a preset sensing target every designated period T (S120). If the power data measured by the sensing step (S120) is out of the threshold range of the initial value (S130), and the power data measured in the comparison step (S130) is out of the threshold range of the initial value (S140) Ex) It is determined that the corresponding power data should be transmitted to the power big data analysis unit 3, and the corresponding power data is transmitted to the power big data analysis unit 3 (S150).

However, if the measured power data is included within a threshold range of the initial value (NO in S140), it is determined that the corresponding power data is not to be transmitted to the power big data analysis unit 3, and the process returns to the sensing step S120. To proceed.

6 is an exemplary view showing a schematic configuration of the power big data analysis unit in FIG. 1.

As shown in FIG. 6, the power big data analysis unit 3 includes a control unit 30, a database server 31, a communication interface unit 32, a power data update unit 33, and a statistical data generation unit ( 34), a data analysis unit 35, a prediction analysis unit 36, a dashboard management unit 37, and a learning unit 38.

The control unit 30 functions as an O.S (Operating System) of the power big data analysis unit 3, and manages and controls control objects (eg, 31, 32, 33, 34, 35, 36, 37).

In addition, the control unit 30 inputs the power data received from the IoT sensors 5 through the communication interface unit 32 to the power data update unit 33 and the statistical data generation unit 34.

In addition, the control unit 30 drives the power data update unit 33 every specified period (T).

In addition, when receiving the reset initial value from the IoT sensor 5, the control unit 30 stores the received reset initial value in the database server 31.

The database server 31 includes identification information of each IoT sensor 5, related information for each power generation facility 4, initial values (or reset initial values) and threshold ranges of each IoT sensor 5, and , Power data received from each IoT sensor 5 and power data replaced (updated) by the power data update unit 33, at least one of a preset analysis algorithm, a machine learning algorithm, and a pattern detection algorithm This is saved.

The operation of the power data update unit 33 will be described in more detail with reference to FIG. 7. 7 is an exemplary view showing a more specific configuration of the power data update unit in FIG. 6.

As illustrated in FIG. 7, the power data update unit 33 is an IoT sensor 5 that does not transmit power data during a designated period T of each IoT sensor 5 (ie, an inactive IoT sensor). Search module 331 for searching for the existence of, and when the inactive IoT sensor is detected by the search module 331, extracts the initial value (or reset initial value) of the inactive IoT sensor detected from the database server 31. It includes an extraction module 332 and a replacement module 333 to replace the initial value (or reset initial value) extracted by the extraction module 332 with power data of the corresponding inactive IoT sensor.

That is, even if the power big data analysis unit 3 according to the present embodiment does not receive power data from the IoT sensors 5, it determines that there is no change from the power data received in the previous cycle and uses the initial value as the power data. By automatically substituting, it is possible to effectively improve network load (overload) due to massive data transmission and reception.

The statistical data generation unit 34 utilizes power data transmitted from each IoT sensor 5 through the communication interface unit 32 and power data generated (updated) by the power data update unit 33. To generate time series statistical data.

The data analysis unit 35 uses the selected or registered analysis algorithm when any one of the preset analysis algorithms is selected from the user (user), or when an analysis algorithm manufactured by the user (user) is registered. Therefore, the time series statistical data for a predetermined period is analyzed.

The operation of the predictive analysis unit 36 will be described in more detail with reference to FIG. 8. 8 is an exemplary view showing a more specific configuration of the prediction analysis unit in FIG. 6.

As shown in FIG. 8, the prediction analysis unit 36 uses a preset pattern detection algorithm to analyze the time series statistical data for a preset period to detect a pattern of each power generation facility 361 And, the similarity detection module 362 for detecting the similarity between the pattern detected by the pattern detection module 361 and the event data learned by the learning unit 38 described later, and the similarity detection module 362 And a prediction module 363 predicting that the corresponding event will occur when the similarity is greater than or equal to the threshold.

The dashboard management unit 37 is a dashboard that outputs time-series statistical data or analysis data generated according to the request of the client 9 (dashboard, to manage and find various information intensively on a single screen on the web) User interface screen).

The learning unit 38 takes power data as an input value, and learns a power event prediction pattern by using a preset machine learning algorithm aimed at power prediction.

As described above, in the present embodiment, when the power data measured by the IoT sensor is within a threshold range of the initial value compared to a preset initial value, the corresponding power data is not transmitted to the power big data analysis unit, and accordingly The big data analysis unit 3 replaces the initial value of the IoT sensor with the power data of the corresponding period when the IoT sensor that has not transmitted the power data is detected during the designated period T, that is, unnecessary power data (ie , It is possible to minimize the network load (overload) by increasing the efficiency of data transmission by not transmitting the same power data as the previous period.

In addition, the present embodiment resets the initial value to the average value of the power data when the power data measured for the preset number of times each IoT sensor is out of the threshold range of the initial value and the difference values of the power data are all within the threshold range ( That is, when the deviation of the power data is within a preset threshold range, the network load can be more efficiently reduced by resetting the preset initial value to the average value of the power data).

The present invention has been described above with reference to the embodiment shown in the drawings, but this is only exemplary, and those skilled in the art to which the art pertains may have various modifications and other equivalent embodiments. You will understand the point. Therefore, the technical protection scope of the present invention should be defined by the following claims.

1: Power time series data transmission processing system
3: Electric Power Big Data Analysis Unit 30: Control Unit
31: database server 32: communication interface unit
33: power data update unit 331: search module
332: extraction module 333: replacement module
34: statistical data generation unit 35: data analysis unit
36: predictive analysis unit 361: pattern detection module
362: similarity detection module 363: prediction module
37: dashboard management unit 38: learning unit
4: Power generation equipment 5: IoT sensor
51: control module 511: sensing module
513: Comparison module 515: Judgment module
517: Reset module 53: Communication module
55: memory 9: client
10: communication network

Claims (15)

At least one Internet of Things (IoT) sensor module unit that measures the state of the power generation facility, analyzes the measurement data, determines whether to transmit it, and then transmits the measurement data according to the determination result;
A power big data analysis unit that receives and analyzes power data from the at least one IoT sensor and generates time series statistical data for each detection target of the power generation facility;
A client accessing the power big data analysis unit to request data, and receiving and monitoring result data corresponding thereto; And
And a communication network connecting the IoT sensor, the power big data analysis unit, and the client by communication.
According to claim 1, The IoT sensor module,
Position sensor, physical sensor, environmental sensor, quality sensor, and fire sensor,
In addition, the transmission processing system for power time series data, characterized in that it comprises at least one of physical and chemical sensors for detecting a state related to the operation of the power generation facility.
According to claim 1, The IoT sensor module,
The initial value and threshold range according to the characteristics of the detection target are preset and stored,
A transmission processing system for power time series data, characterized by determining whether to transmit the power data by analyzing the collected power data using the initial value and the threshold range.
According to claim 3,
The initial value is,
Means a normal measurement value measured at a corresponding sensing target of the power generation facility,
The critical range,
A transmission processing system for power time series data, characterized in that it means a range of variation that can be judged to have not changed from the initial value.
According to claim 1, The IoT sensor module,
When it is determined that the transmission of the power data measured by the power generation facility is meaningful, the power data is transmitted, and when it is determined that the transmission of the power data is meaningless, the power data is not transmitted.
When it is determined that the transmission of the power data is meaningful, it means that the power data is outside a predetermined threshold range,
When it is determined that the transmission of the power data is meaningless, the transmission processing system of the power time series data, characterized in that it means that the power data does not exceed a predetermined take-over range.
According to claim 1, The power big data analysis unit,
Stores a preset initial value and threshold range of the IoT sensor module,
It searches for whether power data has been transmitted from the IoT sensor module every predetermined period (T), and if power data has not been transmitted, it is determined that the power data of the IoT sensor module is included in the threshold range of the initial value, and Power time series data transmission processing system characterized in that the power data of the IoT sensor module is determined as an initial value.
According to claim 1, The power big data analysis unit,
Taking the transmitted power data as an input value, learning the input power data using a preset machine learning algorithm aiming at short-term power prediction,
Predicting and diagnosing short-term power by analyzing power data transmitted from the IoT sensor module using the learned prediction pattern,
When the prediction result comes out, the transmission processing system for power time series data, characterized in that implemented to transmit the prediction data to the client.
According to claim 1, The IoT sensor module,
A communication module that transmits and receives data to and from the power big data analysis unit through a communication network;
A memory for storing an initial value and a threshold range corresponding to the characteristics of the sensing target; And
A transmission processing system for power time series data, comprising; a control module for controlling sensing and transmission of sensed power data according to a specified method.
The method of claim 8, wherein the control module,
A sensing module that measures a state of a preset detection target every period T;
A comparison module that is driven when power data is measured by the sensing module and compares whether the measured power data is included within a threshold range of an initial value stored in the memory; And
When driving the comparison module, if the measured power data is included within a threshold range of the initial value, it is determined that the corresponding power data is not transmitted, and if the measured power data is outside the threshold range of the initial value, the corresponding power data A determination module for determining that it is necessary to transmit; a transmission processing system for power time series data comprising a.
10. The method of claim 9, The control module,
When the power data measured by the sensing module for a predetermined number of times deviates from a threshold range from the initial value, and when the deviations of the power data are values within a predetermined threshold range, the preset initial value is the power. And a reset module for resetting to the average value of the data.
The method of claim 10,
The initial value of the corresponding IoT sensor module reset by the reset module,
The transmission processing system for power time series data, which is transmitted to a power big data analysis unit and resets the received reset initial value to an initial value of the corresponding IoT sensor module.
According to claim 1, The power big data analysis unit,
A communication interface unit that receives power data of the IoT sensor module;
Identification information of IoT sensor module, information about each power generation facility, initial value and threshold range of IoT sensor module, power data received from IoT sensor module, power data updated by power data update unit, preset analysis algorithm, machine A database server storing at least one of a running algorithm and a pattern detection algorithm;
A power data update unit that replaces or updates a preset initial value or a reset initial value when power data is not transmitted from the IoT sensor module;
A statistical data generation unit that generates time series statistical data by utilizing the power data received from the IoT sensor module and the power data updated by the power data update unit;
A data analysis unit for analyzing time-series statistical data for a predetermined period using an analysis algorithm selected or registered by the client;
A dashboard management unit managing a dashboard that outputs time series statistical data or analysis data generated according to the client's request; And
A transmission processing system for power time series data, comprising: a learning unit that learns a power event prediction pattern using a preset machine learning algorithm that targets power prediction using power data as an input value.
The method of claim 12, wherein the power data update unit,
A search module for searching whether there is an inactive IoT sensor module that has not transmitted power data during a designated period T of the IoT sensor modules;
An extraction module extracting an initial value or a reset initial value of the inactive IoT sensor module detected from the database server when the inactive IoT sensor module is detected by the search module; And
And a replacement module that replaces the initial value or the reset initial value extracted by the extraction module with the power data of the corresponding inactive IoT sensor module.
The method of claim 12, wherein the predictive analysis unit,
A pattern detection module that detects a pattern of each power generation facility by analyzing time-series statistical data for a predetermined period using a preset pattern detection algorithm;
A similarity detection module that detects the similarity between the pattern detected by the pattern detection module and the event data learned by the learning unit; And
And a prediction module for predicting that a corresponding event will occur when the similarity is greater than or equal to a threshold value by the similarity detection module.
A method for transmitting power time series data using the power time series data transmission processing system of claim 1,
Pre-setting an initial value and a threshold range of the detection target in the IoT sensor module;
Measuring, by the IoT sensor module, power data of a preset sensing target every designated period (T);
Comparing, by the IoT sensor module, whether the measured power data is outside a threshold range of the initial value;
Determining that it is necessary to transmit the power data to the power big data analysis unit when the measured power data out of the threshold range of the initial value in the comparing step of the IoT sensor module; And
And determining that the power data is not transmitted to the power big data analysis unit when the measured power data is included within a threshold range of the initial value.

Images