AU2020103212A4 - IoT AND MACHINE LEARNING-BASED POWER DISTRIBUTION MANAGEMENT SYSTEM - Google Patents
IoT AND MACHINE LEARNING-BASED POWER DISTRIBUTION MANAGEMENT SYSTEM Download PDFInfo
- Publication number
- AU2020103212A4 AU2020103212A4 AU2020103212A AU2020103212A AU2020103212A4 AU 2020103212 A4 AU2020103212 A4 AU 2020103212A4 AU 2020103212 A AU2020103212 A AU 2020103212A AU 2020103212 A AU2020103212 A AU 2020103212A AU 2020103212 A4 AU2020103212 A4 AU 2020103212A4
- Authority
- AU
- Australia
- Prior art keywords
- tms
- transformer
- machine learning
- main
- fault
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
- G05B23/0275—Fault isolation and identification, e.g. classify fault; estimate cause or root of failure
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/048—Monitoring; Safety
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H5/00—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal non-electric working conditions with or without subsequent reconnection
- H02H5/04—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal non-electric working conditions with or without subsequent reconnection responsive to abnormal temperature
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H7/00—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions
- H02H7/04—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions for transformers
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/001—Methods to deal with contingencies, e.g. abnormalities, faults or failures
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0224—Process history based detection method, e.g. whereby history implies the availability of large amounts of data
- G05B23/0227—Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions
- G05B23/0235—Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
- G05B23/0283—Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Protection Of Transformers (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
"IoT AND MACHINE LEARNING-BASED POWER DISTRIBUTION
MANAGEMENT SYSTEM"
Exemplary aspects of the present disclosure are directed towards the IoT and Machine Learning
Based Power Distribution Management System consists of the plurality of Transformer
Management System (TMS) 101 connected to Smart Power Distribution System (SPDS-MAIN)
103 through Communication Network 102. Transformer Management System (TMS) 101 is an
integration of low-cost microcontroller 101a with advanced power management chip 101b, six
potential transformers 101d, six current sensors 101c. Circuit Breaker 10le, Temperature sensor
101f, and vibrations sensor 101g. TMS 101 uses the Random Forest-based Machine Learning
(ML) algorithm to determine internal and external Faults and operates circuit bakers 101e. All the
power data of the plurality of Transformer Management System (TMS) 101 transmitted to SPDS
MAIN 103 for analyzing the data to identify and predict faults, load variations, transformer
efficiency, losses, and oil temperature variation and isolates the section.
101a 101b
101c
MICRO ADE7816 CT2
CONTROLLER
CT6
CIRCUIT
BREAKERS 101f
101 TRANSFORMER MONITORING SYSTEM (TMS)
Description
101a 101b 101c
MICRO ADE7816 CT2 CONTROLLER
CT6
CIRCUIT BREAKERS 101f
101 TRANSFORMER MONITORING SYSTEM (TMS)
IoT AND MACHINE LEARNING-BASED POWER DISTRIBUTION MANAGEMENT SYSTEM
The following specification particularly describes the invention and the manner in which it is to be performed
[0001] The present disclosure generally relates to the field of Electrical Engineering, more precisely monitoring, protection, Power management and method thereof for a transformer and distribution system.
[0002] In the field of electrical engineering, transformer plays a vital role, and in the process of Electrical Automation, transformer monitoring and protection is a major constrain. Transformer protection is made mainly through a differential protection system, under and over voltage protection system, temperature base protection system, and vibrational base protection system. All the types of protection systems realized with relays, sensors, and microcontroller interface, but they have reasonable failures of the system.
[0003] Though several inventors and researchers have overcome the disadvantages by implementing machine learning and other artificial intelligence-based algorithms in existing microcontroller architecture, but wasn't able to integrate all the protection systems efficiently and unable to integrate with total power distribution system.
[0004] Numerous prior arts have made attempts to automate the grid with numerous prototyping but haven't achieved a more desirable feature in a single unit for the power distribution network.
[0005] Similarly, several prior art disclosures have ascertained that power distribution network reliability depends on power transformer management system. If a transformer is well maintained and able to communicate efficiently then the reliability and efficacy increases drastically.
[0006] In the prior art by Hamid et al. with title Machine Learning Technique as a New Method for Power Transformer Protection, described new classification scheme for power transformer which effectively discriminates between internal faults (turn-to-turns short) and non-intemal faults (magnetizing inrush, external failure and normal condition). Simulation data set of more than 2500 operating states of the transformer have been used in LIBSVM for implementation and validation of the proposed scheme. To illustrate the effectiveness of the proposed scheme in terms of classification accuracy, one type of kernel function of the Support Vector Machine (SVM) classifier has been used. The overall fault classification accuracy has been obtained in which only 40% of the total data (2543) has been used for training, whereas the remaining data (60%) has been applied for testing. Finally, the use of Support Vector Machine (SVM) to discriminate between internal faults and non-internal faults, gives high fault classification accuracy of 97.66%.
[0007] Another prior art by Ravi Shankar Chauhan, titled Internal Fault Detection in Three-Phase Transformer using Machine Learning Methods, presented a study of ABC (artificial bee colony algorithm) and four different machine learning methods has been explored for internal fault detection in three-phase transformer using differential protection scheme. The training and testing result shows that the random forest method gives the best result as compared to the decision tree, linear model, and support vector method.
[0008] Another Prior art document by Sinan Bashi, titled protection of power transformer using microcontroller-based relay - explained paper describes the design and implementation of the microcontroller-based system for protecting power transformer. The system includes facilities for discrimination between intimal fault current and magnetizing inrush current, differential protection, over current protection, overvoltage protection, and under-voltage protection. And also, software and hardware of the microcontroller-based system have been explained and designed. The design implementation and testing of the system also presented.
[0009] Similar prior art by Kajal Salunkhe et al.- discloses Transformer Protection Using Microcontroller Based Relay & Monitoring Using GSM Technology. Overload and overheating protection established for the protection of the transformer. Microcontroller based relay used for the protection of the transformer. The simulation circuit is designed in proteus software, and programming is done in Keil software. In this research, hardware and software of microcontroller-based relay have been explained and designed.
[0010] In Prior art document by Benhmed et al., 2014 with Cost-effective title assessment of transformers using machine learning approach emphasized the ranges of furan content in a power transformer is predicted using measurements of transformer oil tests like breakdown voltage, acidity, and water content. The machine learning approach is adopted, and maintenance data collected from 90 transformers are used. A maximum of 67% recognition rate was achieved using Decision Tree classifier
[0011] CN104919380B discloses a system for the oil temperature of the transformer can be estimated or predicted for the expected load. In this way, it can support the expected load and transformer that can help how long the predicted load is anticipated in the transformer before oil temperature reaches specified threshold value and/or before the transformer is due to the load failure.
[0012] An prior art document by Soto et al., 2019 titled Incipient Fault Diagnosis in Power Transformers by DGA using a Machine Learning ANN-Mean Shift Approach, discussed Dissolved Gas Analysis (DGA) is the best-preferred technique to the diagnosis of incipient faults in power transformers. And also presented an approach to diagnosis fault by DGA using deep neural network, the drawbacks of the number of training patterns (amount of data) is satisfactorily solved with using the Mean Shift algorithm. Likewise, the input and output parameters are conveniently selected, the input parameters being the gas relations established in the IEC 60599 standard acting in parallel with a new ratio of proposed gas (Rnew=C2H2 / C2H6) and binary output. The proposed approach achieved an accuracy of 100%, both in the training and validation process as well.
[0013] Another prior art document US20170228479A1 describes a Systems and methods for real-time modeling of a microgrid are disclosed. An analytics server communicatively connected to a microgrid. The analytics server comprises a virtual system modeling module, an analytics module; a simulation module; and a communications module. The virtual system modeling module is operable to generate predicted data for the microgrid utilizing a virtual system model of the microgrid. The analytics module is operable to initiate a calibration and synchronization operation to update the virtual system model based on a difference between the predicted data and real-time data from the microgrid. The simulation module is operable to forecast at least one aspect of the microgrid. The communications module is operable to provide at least one forecasted aspect to a microgrid operator and a microgrid operator.
[0014] Another prior art document US8751421B2 by Roger N. Anderson Et al disclosed an invention Machine learning for power grid wherein he proposed A machine learning system for ranking a collection of filtered propensity to failure metrics of like components within an electrical grid that includes a raw data assembly to provide raw data representative of the like components within the electrical grid; (b) a data processor, operatively coupled to the raw data assembly, to convert the raw data to more uniform data via one or more data processing techniques; (c) a database, operatively coupled to the data processor, to store the more uniform data; (d) a machine learning engine, operatively coupled to the database, to provide a collection of propensity to failure metrics for the like components; (e) an evaluation engine, operatively coupled to the machine learning engine, to detect and remove non-complying metrics from the collection of propensity to failure metrics and to provide the collection of filtered propensity to failure metrics; and (f) a decision support application, operatively coupled to the evaluation engine, configured to display a ranking of the collection of filtered propensity to failure metrics of like components within the electrical grid.
[0015] Referring to another document, by Mlaki6et al., 2018 discloses A method of Deep learning method and infrared imaging as a tool for transformer faults detection.
[0016] CN102570392A document titled method for identifying exciting inrush current of transformer based on improved probability neural network described the invention relates to a method for identifying an exciting inrush current of a transformer-based on an improved probability neural network. The method comprises the following steps of optimizing smooth factors of the probability neural network by using a genetic algorithm; simulating a model which is constructed in Matlab/Simulink to obtain current waveform; and by taking the wavelet transformation energy of the exciting inrush current and an internal failure current as network input, identifying a failure mode. The method has the advantage that: by applying an intelligent technology to the determination of the exciting inrush current, the failure identification capacity of the exciting inrush current is greatly improved.
[0017] CN103177288A, presented an invention relates transformer fault diagnosis method based on a genetic algorithm optimization neural network. The method includes the steps: 1) selecting H2, CH4, C2H4, C2H6 and C2H2 as fault characteristic gases, and using component contents of the five gases as input vectors of the neural network; 2) using fault free, medium-and-low-temperature overheating, high-temperature overheating, low-energy discharging and high-energy discharging situations as output neurons of the neural network, wherein if an output value is 1, fault diagnosis result belongs to one of the five situations, the larger the value is, the more likely the fault diagnosis result belongs to the five situations, and if the output value is 0, the fault diagnosis result does not belong to the five situations; 3) taking normalized ratios of H2/CH4, CH4/H2 and C2H6/C2H4 as input vectors; 4) selecting an activation function; 5) determining implicit layer number and neuron number of the neural network; and 6) training the neural network. By the transformer fault diagnosis method, convergence speed of the neural network can be effectively increased, convergence precision of the neural network can be effectively improved, and success rate, speed and accuracy rate of fault diagnosis are increased.
[0018] In an early document CN102779230B , The method disclosed by the invention dynamically updates the monitoring data of the transformer through the bayes formula to reflect the state of the transformer in real-time and fully utilizes the historical data information to scientifically calculate the alarm threshold value to figure out the maintenance decision proposal finally so as to provide an auxiliary proposal to production personnel for reasonably arranging the production plan.
[0019] In a prior document CN102221655A, The invention discloses a random-forest model-based power transformer fault diagnosis method, which comprises the following steps of: acquiring transformer state overhauling data, training a random forest model by utilizing the transformer state overhauling state, checking the sensitivity of the random forest model, and diagnosing a fault of a transformer by using the trained and checked random forest model. The method provided by the invention is high in adaptability and interpretability; and criticality between normality and the fault is separated by utilizing a k-means clustering method, and a system is endowed with fault early warning capability.
[0020] In an invention stated in a document US20160358106A1, disclosed a computing device predicts a probability of a transformer failure. An analysis type indicator defined by a user is received. Worth value for each of a plurality of variables is computed. The highest worth variables from the plurality of variables are selected based on the computed worth values. A number of variables of the highest worth variables is limited to a predetermined number based on the received analysis type indicator. A first model and a second model are also selected based on the received analysis type indicator. Historical electrical system data is partitioned into a training dataset and a validation dataset that are used to train and validate, respectively, the first model and the second model. A probability of failure model is selected as the first model or the second model based on a comparison between a fit of each model.
[0021] The present invention provides an effective maintenance system for a Power Distribution System with the integration of the plurality of the Transformer Monitoring System (TMS). Whereas TMS can work alone with Machine Learning Algorithm and as well as in coordination control with other plurality of TMS and SPDS-MAIN. Wherein SPDS MAIN on a whole can execute advanced Machine Learning Algorithms to ascertain stress on the lines and isolate the faulty or overloaded systems.
[0022] The present invention addresses the shortcomings mentioned above of the prior art.
[0023] All publications herein are incorporated by reference to the same extent as if each publication or patent application were specifically and individually indicated to be incorporated by reference. Where a definition or use of a term in an incorporated reference is inconsistent or contrary to the definition of that term provided herein, the definition of that term provided herein applies, and the definition of that term in the reference does not apply.
[0024] The following presents a simplified summary of the disclosure in order to provide a basic understanding of the reader. This summary is not an extensive overview of the disclosure, and it does not identify key/critical elements of the invention or delineate the scope of the invention. Its sole purpose is to present some concepts disclosed herein in a simplified form as a prelude to the more detailed description that is presented later.
[0025] Exemplary embodiments of the present disclosure are directed towards the IoT and Machine Learning-Based Power Distribution Management System.
[0026] An exemplary object of the present disclosure is directed towards a system that monitors and controls the transformer.
[0027] Another exemplary object of the present disclosure is directed towards the integration of microcontroller 101a with Power IC 101b to mark Transformer Management System (TMS) 101. Whose primary function is to monitor vital transformer parameters especially voltages, currents, frequency, vibration and transformer core and oil temperatures.
[0028] Another exemplary object of the present disclosure is directed towards the integration of microcontroller 101a with a vibrations sensor 101g and thermal camera 101f for identifying core temperature variations and vibrational analysis and oil.
[0029] An exemplary aspect of the present subject matter is directed towards the implementation of the Machine Learning Algorithm in microcontroller 10la for ascertaining the type of internal fault that occurred in the transformer.
[0030] An exemplary aspect of the present subject matter is directed towards the use of the Machine Learning Algorithm in microcontroller 101a for predicting the fault event that may be caused by excess vibrations of the transformer core.
[0031] An exemplary aspect of the present subject matter is directed towards the implementation of the Machine Learning Algorithm in microcontroller 10la for predicting the fault event that may be caused by the excessive temperature rise of the transformer oil or core.
[0032] Another exemplary aspect of the present disclosure is directed towards the integration of the plurality of TMS with Power Distribution Management System (PDMS) MAIN through WiFi Mesh Network. Where in each TMS acts as a node and trans-receive the WiFi signals.
[0033] Another exemplary aspect of the present disclosure is directed towards the implementation of the Machine Learning Algorithm in PDMS-MAIN for supporting decisions of the plurality of TMS during fault conditions and prediction of failures.
[0034] Another exemplary aspect of the present disclosure directed towards the implementation of the Machine Learning Algorithm in SPDS-MAIN for determining the stress on line and loading capacity.
[0035] Another exemplary aspect of the present disclosure directed towards the removal of the faulty section or transformer from the system to avert total system failure.
[0036] In the following, numerous specific details are set forth to provide a thorough description of various embodiments. Certain embodiments may be practiced without these specific details or with some variations in detail. In some instances, certain features are described in less detail so as not to obscure other aspects. The level of detail associated with each of the elements or features should not be construed to qualify the novelty or importance of one feature over the others.
[0037] FIG.1 is a diagram depicting the100 SMART POWER DISTRIBUTION MANAGEMENT SYSTEM, according to an exemplary embodiment of the present disclosure.
[0038] FIG. 2 is a Block diagram 101 TRANSFORMER MONITORING SYSTEM (TMS) according to an exemplary embodiment of the present disclosure.
[0039] FIG. 3 is an actual representation of the microcontroller 101a, according to an exemplary embodiment of the present disclosure.
[0040] FIG. 4 is an actual representation of Custom Made PCB for POWER IC (ADE7861), according to an exemplary embodiment of the present disclosure.
[0041] FIG. 5 is a flow chart 500 depicting the Process for Internal Fault determination in TMS.
[0042] FIG. 6 is a flow chart 600 depicting the process carried in SPDS-MAIN.
[0043] It is to be understood that the present disclosure is not limited in its application to the details of construction and the arrangement of components outlined in the following description or illustrated in the drawings. The present disclosure is capable of other embodiments and of being practiced or of being carried out in various ways. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting.
[0044] The use of "including", "comprising" or "having" and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items. The terms "a" and "an" herein do not denote a limitation of quantity, but rather denote the presence of at least one of the referenced item. Further, the use of terms "first", "second", and "third", and the like, herein do not denote any order, quantity, or importance, but rather are used to distinguish one element from another.
[0045] The nomenclature used hereafter for ease of understanding grouped and termed here as follows. Power distribution system means a plurality of power transformers connected in a ring to maintain continuous supply to all the loads by deriving the power from one or more power sources. Microcontroller means a microcomputer capable of executing Artificial intelligence and machine language algorithms. Machine Learning Algorithm (MLA) implies a generalized term for the complete process of data collections, noise removal, data formatting, cleansing, usage of random forest method, training the model with 20-80 ration, model evaluation, parameter tuning, prediction, and weight adjustment. Power IC 101b is an advanced chip custom made to collect power data in a transformer having a six-channel current sensor in which three connected in input/primary side and another three connected to secondary. Another set of nomenclature used hereafter are defined when and where required.
[0046] Referring to FIG. 1 is a diagram depicting an IoT and Machine Learning Based Distribution System 100. Which comprises of a SMART POWER DISTRIBUTION MANAGEMENT SYSTEM, a plurality of Transformer Monitoring System (TMS) 101 connected through WiFi Mesh network 102 to SMART POWER DISTRIBUTION SYSTEM (SPDS)-MAIN 103. SPDS-MAIN is a central processing unit capable of executing parallel and complex Machine Learning Algorithms (MLA) and calculations. All the TMS 101 are enabled with WiFi mesh network which eliminates the necessity of any special routers or modems and this feature is an advanced IoT function that removes the redundancy and reliable data transfer.
[0047] Further to it, SPDS-MAIN uses a trained MLA based on a random forest model for predicting faults, line loading, load flow analysis, and fault determination. Data transmitted by the plurality of TMS 101 is collected, labelled as per TMS ID and time interval, and stored in the database. The power data thus received is cleansed and fed to the MLA for predicting stress on the feeder and line, the capacity of other components in the line. And also, MLA predicts the faults and analyzes the fault events which take place in or outside the transformer and justifies the actions taken by TMS 101.
[0048] Additional to it, SPDS-MAIN can isolate a faulty section or an overstressed section based on the prediction model. This feature enables the distribution system to maintain its stability and reliability through its operation time.
[0049] In accordance with a non-limiting exemplary embodiment of the present subject matter, FIG. 2 is a representation of TRANSFORMER MONITORING SYSTEM (TMS) 101. As depicted in the drawings of FIG 2, TMS 101 consists of microcontroller 101a, Power IC (ADE7816) 101b, Current sensors 101c, Potential sensors 101d, embedded connection to Circuit breakers 101e controlled through relays, Temperature sensor 101f and vibrational sensor 101g. Current sensors 101c divided into a set of three and connected in primary and secondary of the transformer. Similarly, Potential transformers 101d divided into a set of three and connected in primary and secondary of the transformer. The arrangement of Current sensors 101c forms a differential protection system for identifying both internal, external faults as well as inrush and overcurrent's. The combinational performance analysis of temperature sensor 101f and vibration sensor 101g enables the relevant MLA to predict the future faults arising of core vibrations and oil temperature variations with respect to the predicted and present load. Microcontroller 10la controls the circuit breakers concerned with that transformer, which are arranged in incoming and outgoing lines, as indicated in FIG 1. Based on the output of MLA, Microcontroller 101a decided to operate concerned circuit breaker 101e for averting faults and disconnecting the faulty transformer. All the components of the TMS concealed in a custom container with IP & IK protection norms and for ease of installation. Powered by solar, and as well as +12/+5 V DC supply, makes the TMS more convenient for deployment.
[0050] Referring to FIG 3 is a diagram depicting the actual image of Microcontroller 101a, which acts as a backbone for the present invention. Tweakedwith ZRAM and additional external storage system 101a-i makes the microcontroller 101a suitable for the present invention. In microcontroller 101a, Machine Learning Algorithms can be executed in parallel to validate several parameters of the transformer.
[0051] In accordance with a non-limiting exemplary embodiment of the present subject matter, FIG. 4 is an actual image depicting the Custom Made PCB for POWER IC(ADE7816) 101b. The Power IC 101b- is a state of the art chip designed particularly for accurate power measurement and integrated with easy. By utilizing this feature, in the present invention, the same Power IC 10lb-i is developed on a custom size PCB to hold the required components such as power supply and sensor connectors and controller connectors. Custom PCB integrated with microcontroller 10la for transformer protection. The Current Transformers 101c, which are split-core type connected to any sort of cables/wire up to 200A current rating. Three current transformers 101c form a group that is hooked up in either the primary or secondary side of the transformer and acts as a differential protection scheme. Power IC 101b-i collects the sensor data and accurately measures and presents power data, which are Active Power, Reactive Power, Apparent Power, Individual phase and line currents and voltages, along with RMS and True values and concerned waveform.
[0052] According to a non-limiting exemplary embodiment of the present disclosure, FIG.5 is a flow chart 500 depicting Process for Internal Fault determination in TMS. TMS considered as principal protection devises for the transformer in the way it acts by monitoring its performance with sensors and power IC and controlling its operations when fault or abnormality arises. The step 501 starts by Acquire Current magnitudes and Voltage magnitudes from Power IC 101b and assign a time-stamp to the acquired values along with TMS ID. Timestamping is done to the acquired value so that the occurrence can be known precisely for the present and future process. Whereas TMS ID is attached to the data acquired as this data is further transmitted to SPDS-MAIN for justification and prediction of the faults and other events, and this TMS ID makes the SPDMS-MAIN to pinpoint the occurrence point in the system accurately. In preceding step 502, microcontroller 101a compare magnitudes of input/output currents and voltages of the transformer and compute the change of currents and voltages Al and AV. In a balanced system, when differential protection is employed, then the Al and AV must be nearly zero. Therefore, in step 503 Ascertain, whether Al and AV values are in limits by comparing them with set values. For speedy of response, the microcontroller first compares the Al and AV values with user defined set values to determine an ongoing fault. In step 504, If values are persistently high but not matching with set values then, feed the values directly to the machine learning algorithm and predict the type of fault, alert the SPDS-MAIN, and trip the relay for opening the concerned circuit breaker 101e.
[0053] Further to step 505, which is a wait cycle, wherein microcontroller 101a after two cycles of power frequency energies the transformer. This step is crucial for eliminating any false fault indication arisen from several constraints. Next step 506 starts with Acquiring Al and AV after energies the transformer by turning on circuit breaker 101e of the concerned transformer. And in the following step 507, microcontroller 10a ascertain that Al is a fault current or an inrush current after circuit breaker re-closes with the help of a machine learning algorithm. MLA, with a trained model for predicting the difference between fault current or an inrush current, is employed to ascertain the fault and the action to take presented in the next step 508. Wherein step 508, a conclusion about the fault has arrived from 507, and the microcontroller 101a recorders the data completely shut the transformer and intimate the SPDS-MAIN with time stamping & TMS ID and upload power data and type of fault, thereby ends the process 500.
[0054] According to a non-limiting exemplary embodiment of the present disclosure, FIG.6 is a flow chart 600 depicting the process carried in SPDS-MAIN, which starts at step 601 wherein SPDS-MAIN acquire all the parameters of Transformers through respective sensors and assign time-stamping along with TMS ID. Data is not restricted to sensors, but it also includes the MLA predicted outcome as well as other derivatives. As data collected from the plurality of TMS, then the data cleansing should take place. Therefore in step 602. Data cleansed to meet the requirements of the MLA present in SPDS-MAIN. In preceding step 603, data of Al, AV, AT, and APL of all the plurality of TMS fed to the concerned machine learning algorithm MLA. Al, AV values are fed to one type of MLA trained for predicting the faults and timeline of the faults. Whereas AT(Temperature variations), Vibrations variation data, and APL(Cre Losses) fed to another type of MLA trained for predicting the faults and timeline of the faults arise due to transformer core temperature and vibrations. In step 604. Relevant MLA is executed on the data to predicts the line losses and line stress in all sections.
[0055] Further, in step 605, with the help of predicted outcome from one of the MLAs concerning current and voltage variations in all the lines, SPDS- MAIN ascertains the constraints and if line losses or line stress is high in a particular section, compare the data of nearby TMS 101 for any anomaly in power data and remove the section by operating concerned circuit breakers 101e and informs the user about the condition. While in step 606,
SPDS-MAIN 103 Monitor the particular TMS 101 due to which anomaly has occurred and if the parameters come within range restore the particular TMS 101 and section and intimate the user. In consecutive step 607, if microcontroller in SPDS-MAIN 103 predicts a fault then shut the line permanently and intimate the user about the problem.
[0055] In an embodiment, SPDS-MAIN 103 collects the data of transformers core vibrations, the temperature of core and oil and currents in each section. This data is normalized and fed to relevant MLA to predict the nearby fault conditions and stress on the line due to excessive currents or losses in a transformer. It is to note that, excessive heat and vibrations lead to transformer fault and as the transformer reaches the fault condition its power losses increase and dramatically increases the fault currents. If predicted results show adverse loading condition and stress on the line, SPDS-MAIN removes the faulty condition.
EDITORIAL NOTE 2020103212 There are two pages of claims only
Claims (6)
1. The IoT and Machine Learning-Based Power Distribution Management System:
A plurality of a Transformer Management System (TMS) 101, and a SPDS-MAIN 103 connected through WiFi Mesh Network 102; and Where in Transformer Management System (STMS) 101 is an integration of Microcontroller 101a, Power IC 101b, Current Sensors 101c, Potential Transformers 101d, circuitry for controlling Circuit Breaker 101e, Temperature sensor 101f and Vibration sensor 101g; and TMS 101 capable of running suitable Machine Learning Algorithms for predicting and ascertaining the type of fault, fault timeline; and TMS 101 capable of running suitable Machine Learning Algorithms for predicting fault due to temperature and core vibrations of the transformer; and SPDS-MAIN is comprising a computer capable of acting as a server and parallel execution of different Machine Learning Algorithms to ascertain decisions taken by TMS; and SPDS-MAIN capable of executing a type of Machine Learning algorithm to predict the loading effect on line and isolating the faulty session based on current statistics in each section. SPDS-MAIN capable of running suitable Machine Learning Algorithms on data acquired from temperature and core vibrations of the all transformer for predicting loading and fault conditions.
2. The device as claimed in claim 1, wherein a combination of a plurality of TMS and SPDS-MAIN for monitoring and controlling the power distribution system. And the communication exists between TMS and SPDS-MAIN is a WiFi mesh network, where each TMS acts as a transceiver. Each TMS acts as a node in a mesh, which means it acts as an access point (AP) and WiFi repeater, thereby removing the necessity of the WIFi router and extends the range to kilometers.
3. The device as claimed in claim 1, wherein the TMS contains the integration of Microcontroller 101a and Power IC (ADE7816) 10lb capable of making a differential protection scheme for the transformer and determine/ predict internal or external faults of a transformer by using suitable Machine Learning Algorithm.
4. The device as claimed in claim 1, wherein the TMS 101 contains the integration of Microcontroller 101a with Temperature sensor 101f and Vibration Sensor 101g for predicting Transformer faults due to excessive core vibrations and heat and oil temperature variations by using suitable Machine Learning Algorithm.
5. The device as claimed in claim 1, wherein SPDS-MAIN using Machine Learning Algorithm, determine the stress on the line before or after a fault and isolates the section.
6. The device as claimed in claim 1, wherein SPDS-MAIN use Machine Learning Algorithm to determine the stress on the line based on transformer core vibrations and heat and as well as oil temperature data received from all the TMS 101 and isolate the section to avoid failure.
103 SPDS -MAIN
101a4 102 101a4 2020103212
101a1
104-n TMS1
101e-1
101e-13 101e-12 101a3
105
104-2
104-3 TMS2 101a2
FIG 1 POWER DISTRIBUTION MANAGEMENT SYSTEM
2020103212 2
101a 101b 101c
CT1
CT2 MICRO ADE7816 CONTROLLER POWER IC
CT6
PT
101d 1-6
101e 101g CIRCUIT BREAKERS 101f
101 TRANSFORMER MONITORING SYSTEM (TMS)
FIG 3 101a-1 101a
2020103212 4
101c
101b-1
FIG 4- Custom Made PCB with POWER IC ADE7816
500 501 04 Nov 2020
Acquire Current magnitudes and Voltage magnitudes from Power IC 101b. Assign time stamping to the acquired values along with TMS ID 502 Compare magnitudes of input/output currents and voltages of transformer and compute the change of currents and voltages ΔI and ΔV.
503 Ascertain whether ΔI and ΔV values are in limits by comparing them with set 2020103212
values and using relevant machine learning algorithm.
504 If ΔI and ΔV are very high, then feed the values directly to machine learning algorithm and predict the type of fault alert the PDMS-MAIN and trip the relay for opening the concerned circuit breaker 101e. 505
Wait for two power cycles
506
Acquire ΔI and ΔV after energies the transformer by turning on circuit breaker 101e of concerned transformer. 507
Ascertain that, ΔI is a fault current or an inrush current after circuit breaker re-closure with the help of machine learning algorithm.
508 If ΔI is high and persistent after the re-closure test, then ascertain the type of fault and completely shut the transformer and intimate the PDMS-MAIN with time stamping & TMS ID and upload power data and type of fault.
FIG 5 500 Process for Internal Fault determination in TMS
600 601
Acquire all the parameters of Transformers through respective sensors and assign time stamping along with TMS ID
602
Segregate the sensor data concerning its type and arrange them in sequence 2020103212
and place of occurrence and perform data cleansing.
603
Feed the cleansed data of ΔI, ΔV, ΔT, and ΔPL of all the plurality of TMS fed to the machine learning algorithms.
604
Machine learning algorithm predicts the line losses and line stress in all sections.
605 If line losses or line stress is high in a particular section, compare the data of nearby TMS 101 and remove the section by operating concerned circuit breakers 101e. 606
Monitor the particular TMS 101 due to which anomaly has occurred and if the parameters come within range restore the particular TMS 101 and section and 607 intimate the user.
Else shut the line permanently and intimate the user about the problem.
FIG 6; 600 Process carried in PDMS-MAIN
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
AU2020103212A AU2020103212A4 (en) | 2020-11-04 | 2020-11-04 | IoT AND MACHINE LEARNING-BASED POWER DISTRIBUTION MANAGEMENT SYSTEM |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
AU2020103212A AU2020103212A4 (en) | 2020-11-04 | 2020-11-04 | IoT AND MACHINE LEARNING-BASED POWER DISTRIBUTION MANAGEMENT SYSTEM |
Publications (1)
Publication Number | Publication Date |
---|---|
AU2020103212A4 true AU2020103212A4 (en) | 2021-01-14 |
Family
ID=74103463
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
AU2020103212A Ceased AU2020103212A4 (en) | 2020-11-04 | 2020-11-04 | IoT AND MACHINE LEARNING-BASED POWER DISTRIBUTION MANAGEMENT SYSTEM |
Country Status (1)
Country | Link |
---|---|
AU (1) | AU2020103212A4 (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114167217A (en) * | 2021-12-09 | 2022-03-11 | 中国路桥工程有限责任公司 | Multiple fault diagnosis method for railway power distribution network |
CN117910120A (en) * | 2024-03-20 | 2024-04-19 | 西华大学 | Buffeting response prediction method for wind-bridge system based on lightweight transducer |
CN118091331A (en) * | 2024-04-26 | 2024-05-28 | 国网辽宁省电力有限公司抚顺供电公司 | Cable fault sensing method and system |
CN118194150B (en) * | 2024-05-13 | 2024-08-02 | 无锡冠亚恒温制冷技术有限公司 | Remote monitoring and fault prediction system for water chiller |
-
2020
- 2020-11-04 AU AU2020103212A patent/AU2020103212A4/en not_active Ceased
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114167217A (en) * | 2021-12-09 | 2022-03-11 | 中国路桥工程有限责任公司 | Multiple fault diagnosis method for railway power distribution network |
CN117910120A (en) * | 2024-03-20 | 2024-04-19 | 西华大学 | Buffeting response prediction method for wind-bridge system based on lightweight transducer |
CN118091331A (en) * | 2024-04-26 | 2024-05-28 | 国网辽宁省电力有限公司抚顺供电公司 | Cable fault sensing method and system |
CN118091331B (en) * | 2024-04-26 | 2024-07-05 | 国网辽宁省电力有限公司抚顺供电公司 | Cable fault sensing method and system |
CN118194150B (en) * | 2024-05-13 | 2024-08-02 | 无锡冠亚恒温制冷技术有限公司 | Remote monitoring and fault prediction system for water chiller |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
AU2020103212A4 (en) | IoT AND MACHINE LEARNING-BASED POWER DISTRIBUTION MANAGEMENT SYSTEM | |
US8200372B2 (en) | Methods and processes for managing distributed resources in electricity power generation and distribution networks | |
US10422833B2 (en) | Systems and methods for classifying power line events | |
RU2313815C2 (en) | Device and method for controlling technical plant, which contains a set of systems, in particular, electric power plant | |
CN116722249A (en) | Battery thermal runaway early warning protection system and protection method thereof | |
RU2649542C1 (en) | Method and system of remote monitoring of objects | |
EP3748811B1 (en) | A method for configuring an intelligent electronic device and a system therof | |
CN108828406A (en) | The fault recognition method and its system of non-intrusion type user power utilization | |
Sowah et al. | Design of power distribution network fault data collector for fault detection, location and classification using machine learning | |
CN115684821A (en) | System and method for AI continuous learning in grid fault analysis | |
CN112734977B (en) | Equipment risk early warning system and algorithm based on Internet of things | |
CN106773993A (en) | Field data acquiring terminal and system | |
Kavana et al. | Fault analysis and predictive maintenance of induction motor using machine learning | |
Kezunovic | Translational knowledge: From collecting data to making decisions in a smart grid | |
CN117674140B (en) | Power distribution network measurement and control system and method | |
CN113219330B (en) | Method and system for detecting state of isolating switch | |
CN117949887A (en) | Fault detection and prevention method for electric energy metering system of energy controller | |
Enns et al. | Potential applications of expert systems to power system protection | |
CN117031201A (en) | Multi-scene topology anomaly identification method and system for power distribution network | |
CN107102633A (en) | A kind of distribution terminal fault self-diagnosis method and system | |
CN110807014A (en) | Cross validation based station data anomaly discrimination method and device | |
CN116192038A (en) | Photovoltaic system fault diagnosis method and device and computer equipment | |
AU2020104355A4 (en) | IoT and MACHINE LEARNING BASED POWER QUALITY IMPROVEMENT SYSTEM FOR MICRO-GRID | |
Provči et al. | A New Protection System Design of Active MV Distribution Network | |
EP4404407A1 (en) | Data processing system and data processing method for a substation system, and electric power system substation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FGI | Letters patent sealed or granted (innovation patent) | ||
MK22 | Patent ceased section 143a(d), or expired - non payment of renewal fee or expiry |