CN111680831A - Substation power utilization analysis system and method - Google Patents
Substation power utilization analysis system and method Download PDFInfo
- Publication number
- CN111680831A CN111680831A CN202010450628.7A CN202010450628A CN111680831A CN 111680831 A CN111680831 A CN 111680831A CN 202010450628 A CN202010450628 A CN 202010450628A CN 111680831 A CN111680831 A CN 111680831A
- Authority
- CN
- China
- Prior art keywords
- power consumption
- station
- electric energy
- substation
- data server
- 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.)
- Pending
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 41
- 238000000034 method Methods 0.000 title abstract description 32
- 238000013528 artificial neural network Methods 0.000 claims abstract description 84
- 238000004891 communication Methods 0.000 claims abstract description 22
- 238000012546 transfer Methods 0.000 claims description 51
- 230000005611 electricity Effects 0.000 claims description 46
- 230000008859 change Effects 0.000 claims description 17
- 238000010606 normalization Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 10
- 238000007405 data analysis Methods 0.000 abstract description 5
- 230000005540 biological transmission Effects 0.000 description 17
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 16
- 230000006870 function Effects 0.000 description 9
- 238000001816 cooling Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 210000002569 neuron Anatomy 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 238000011478 gradient descent method Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 239000007788 liquid Substances 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 239000000498 cooling water Substances 0.000 description 2
- 239000008367 deionised water Substances 0.000 description 2
- 229910021641 deionized water Inorganic materials 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- QVFWZNCVPCJQOP-UHFFFAOYSA-N chloralodol Chemical compound CC(O)(C)CC(C)OC(O)C(Cl)(Cl)Cl QVFWZNCVPCJQOP-UHFFFAOYSA-N 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000001802 infusion Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
- G16Y10/00—Economic sectors
- G16Y10/35—Utilities, e.g. electricity, gas or water
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
- G16Y40/00—IoT characterised by the purpose of the information processing
- G16Y40/20—Analytics; Diagnosis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Biomedical Technology (AREA)
- Development Economics (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Primary Health Care (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Accounting & Taxation (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The application relates to a substation power utilization analysis system and method. The system comprises a front-end sensor and a data server, wherein the front-end sensor and the data server are in communication in a wireless mode, the front-end sensor is used for detecting electric energy parameters of the transformer substation and sending the electric energy parameters to the data server, and the data server is used for analyzing the electric energy parameters based on a BP (back propagation) neural network, establishing a power grid fault early warning model and predicting the load of the transformer substation according to the power grid fault early warning model. The front-end sensor and the data server are communicated in a wireless mode, the use cost can be reduced, communication lines do not need to be laid, the safety of the transformer substation is improved, the electric energy parameters of the data server are subjected to data analysis modeling, a fault early warning model based on the electric energy parameters is obtained, the model can be used for predicting the load of the transformer substation, faults are pre-judged in advance, so that solving measures can be taken in time, and the use reliability of the power utilization analysis system of the transformer substation is improved.
Description
Technical Field
The application relates to the technical field of power systems, in particular to a substation power utilization analysis system and method.
Background
With the development of the power industry, the reliability and the power quality of the transformer substation are more and more concerned by people, and to analyze the reliability of the transformer substation, working parameters, environmental parameters and the like in the transformer substation are collected firstly, then the parameters are analyzed, the working state of the transformer substation is analyzed, and corresponding processing is performed to improve the power quality, so that the collection of power grid data is the basis for improving the power quality.
In the traditional power grid data acquisition, a wired acquisition scheme is generally adopted, power grid data are transmitted to a data processing center through a communication line, and the data processing center analyzes the received data and returns corresponding instructions to all devices of a transformer substation. However, the cost of the wired transmission scheme is high, especially the occupied area of a transformer substation with the voltage of more than 220kV is large, the cost of laying communication lines is high, and because the laying of the communication lines is generally in live working, the working time is long, the working risk is high, and the use reliability is low.
Disclosure of Invention
Based on this, it is necessary to provide a substation power consumption analysis system and method for solving the problem of low reliability of traditional power grid data acquisition and use.
A power utilization analysis system for a transformer substation comprises a front-end sensor and a data server, wherein the front-end sensor and the data server are communicated in a wireless mode;
the front-end sensor is used for detecting electric energy parameters of the transformer substation and sending the electric energy parameters to the data server, and the data server is used for analyzing the electric energy parameters based on a BP neural network, establishing a power grid fault early warning model and predicting the load of the transformer substation according to the power grid fault early warning model.
A substation electricity utilization analysis method comprises the following steps:
acquiring electric energy parameters of a transformer substation detected by a front-end sensor;
analyzing the electric energy parameters based on a BP neural network, and establishing a power grid fault early warning model;
and predicting the load of the transformer substation according to the power grid fault early warning model.
According to the power utilization analysis system and method for the transformer substation, the front-end sensor and the data server are in communication in a wireless mode, the front-end sensor detects the electric energy parameters of the transformer substation and sends the electric energy parameters to the data server, the data server analyzes the electric energy parameters based on the BP neural network, a power grid fault early warning model is built, and the power grid fault early warning model is used for predicting the load of the transformer substation. The front-end sensor and the data server communicate in a wireless mode, the use cost can be reduced, communication lines do not need to be laid, the safety of the transformer substation is improved, the data server can introduce a BP (back propagation) neural network after acquiring electric energy parameters of the transformer substation, data analysis modeling is carried out on the electric energy parameters, a fault early warning model based on the electric energy parameters is obtained, the model can be used for predicting the load of the transformer substation, the faults are pre-judged in advance, so that solution measures can be taken in time, and the use reliability of the transformer substation power consumption analysis system is improved.
In one embodiment, the front-end sensor comprises an infrared meter reader and a collector, the infrared meter reader and the collector are both in wireless communication with the data server, and the electric energy parameters comprise station power consumption and main transformer transfer capacity of a power utilization module in a power utilization station; the infrared meter reading device is used for detecting the station power consumption and sending the station power consumption to the data server, and the collector is used for collecting the main transformer transfer capacity and sending the main transformer transfer capacity to the data server.
In one embodiment, the substation power consumption analysis system further comprises a centralized gateway, wherein one side of the centralized gateway is connected with the front-end sensor, and the other side of the centralized gateway is wirelessly connected with the data server.
In one embodiment, the centralized gateway is wirelessly connected to the data server via a WAPI wireless network.
In one embodiment, the front-end sensor includes an infrared meter reader and a collector, the electric energy parameters include station power consumption and main transformer transfer capacity of a power consumption module in a power station, and the acquiring the electric energy parameters of the substation detected by the front-end sensor includes:
and acquiring the station power consumption detected by the infrared meter reader and the main transformer transfer capacity acquired by the collector.
In one embodiment, the analyzing the electric energy parameter based on the BP neural network, and the establishing a power grid fault early warning model includes:
calculating to obtain station power consumption according to the station power consumption and the main transformer transfer capacity;
calculating to obtain a station electricity utilization change rate according to the station electricity utilization rate and a preset station electricity utilization rate;
and analyzing the station power consumption rate and the station power consumption change rate based on the BP neural network, and establishing a power grid fault early warning model.
In one embodiment, the analyzing the station power consumption rate and the station power consumption change rate based on the BP neural network, and the establishing a power grid fault early warning model includes:
taking the station electricity utilization rate and the station electricity utilization change rate as input information sources of the BP neural network;
acquiring the number of nodes of a hidden layer of the BP neural network;
and training the BP neural network according to an LM algorithm based on the input information source and the node number, and establishing a power grid fault early warning model.
In one embodiment, the calculating the station power consumption according to the station power consumption and the main transformer transfer capacity comprises:
acquiring historical maximum power consumption corresponding to the station power consumption in a preset acquisition period;
taking the historical maximum power consumption as a reference, and carrying out normalization processing on the station power consumption;
and calculating to obtain the station power consumption according to the station power consumption after the normalization processing and the main transformer transfer capacity.
In one embodiment, the acquiring the electric energy parameter of the substation detected by the front-end sensor includes:
network security authentication is carried out on the electric energy parameters of the transformer substation detected by the front-end sensor;
and receiving the electric energy parameter after the electric energy parameter passes the network security authentication.
Drawings
FIG. 1 is a block diagram of a substation power consumption analysis system in one embodiment;
FIG. 2 is a schematic diagram of data acquisition of a substation power consumption analysis system in one embodiment;
FIG. 3 is a block diagram of a substation power consumption analysis system in another embodiment;
FIG. 4 is a flow diagram of a substation power analysis method in one embodiment;
FIG. 5 is a flow chart of a substation power analysis method in another embodiment;
FIG. 6 is a flow chart of a substation power analysis method in yet another embodiment;
FIG. 7 is a flow chart of a substation power analysis method in yet another embodiment;
FIG. 8 is a schematic diagram of security certification of a substation power analysis method in one embodiment;
FIG. 9 is a diagram of WAPI authentication in one embodiment;
FIG. 10 is a diagram illustrating operation of a BP neural network according to one embodiment;
FIG. 11 is a flowchart illustrating the operation of the BP neural network in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described more fully below by way of examples in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In one embodiment, please refer to fig. 1, which provides a power consumption analysis system for a transformer substation, including a front-end sensor and a data server, where the front-end sensor communicates with the data server in a wireless manner, the front-end sensor is configured to detect an electric energy parameter of the transformer substation and send the electric energy parameter to the data server, and the data server is configured to analyze the electric energy parameter based on a BP neural network, establish a power grid fault early warning model, and predict a load of the transformer substation according to the power grid fault early warning model. The front-end sensor and the data server communicate in a wireless mode, the use cost can be reduced, communication lines do not need to be laid, the safety of the transformer substation is improved, the data server can introduce a BP (back propagation) neural network after acquiring electric energy parameters of the transformer substation, data analysis modeling is carried out on the electric energy parameters, a fault early warning model based on the electric energy parameters is obtained, the load of the transformer substation can be predicted, faults are pre-judged in advance, so that solution measures can be taken in time, and the use reliability of the transformer substation power consumption analysis system is improved.
Specifically, the front-end sensor detects electric energy parameters of each power utilization module in the transformer substation and sends the electric energy parameters to the data server, the number and the types of the power utilization modules are not unique, generally speaking, the more the number and the types of the power utilization modules are, the richer the data samples collected by the front-end sensor are, and the more the power grid fault early warning model obtained based on the abundant electric energy parameters is close to the real situation, the more accurate the model is. The number of the front-end sensors is not unique, when the number of the power utilization modules is more than two, a plurality of front-end sensors can be adopted, sensors which are not used correspond to electric energy parameters of different power utilization modules, so that the detection efficiency is improved, or the same front-end sensor can be adopted to detect the electric energy parameters of the power utilization modules, so that the use cost is reduced.
In this embodiment, taking an example that the electricity utilization module includes a STATCOM (Static Synchronous Compensator) system, a transformer and an ac feeder, the front end sensor detects real-time quantity in the STATCOM system, including water supply flow temperature and pressure, water return temperature, air intake temperature, pressure of an inlet of a main circulation pump and a buffer tank, conductivity of cooling water and deionized water, liquid level of the buffer tank, temperature and humidity of a valve hall, and detects working parameters related to states of the main circulation pump, an electric heater, a liquid replenishing pump, an air replenishing solenoid valve and a fan in the STATCOM system and sends the working parameters to the data server, the front end sensor also detects main transformer electricity quantity, oil temperature, high-up current loop temperature, fan electricity weight and rotation speed in the transformer and sends the working parameters to the data server, the front end sensor also detects electricity quantity in a total loop, power, lighting, a dc charger and an office protection room environment in the ac feeder, and temperature and humidity in the transformer substation, The air conditioner electric quantity and the lighting electric quantity are sent to a data server, the data server takes the data as input information sources of a BP neural network, the BP neural network takes the acquired information quantities of a plurality of sources as input signals, and an output value is generated through calculation of the BP neural network.
The BP neural network is a multi-layer feedforward neural network, the model of which comprises an initial input layer, an intermediate hidden layer and a final output layer, and the algorithm of the BP neural network comprises two processes of forward propagation of signals and backward propagation of errors. When the signal is transmitted in the forward direction, the input signal acts on the output node through the middle hidden layer, and the calculated output value is generated through nonlinear transformation. And comparing the calculated output value with an expected output value (the real-time information of the electric energy parameter of the transformer substation), and generating an output error when the calculated output value is different from the expected value, so that the error is transferred to a back propagation process. The error back transmission is to back transmit the output error to the input layer through the middle hidden layer, and distribute the error to all units of each layer, and the distributed error semaphore is used as the basis for adjusting the weight of each unit. And repeatedly iterating the network parameters by using the minimum error as an approaching target through a gradient descent method, thereby determining the network parameters corresponding to the minimum error, recording the adjusted weight, and establishing a fault early warning model by using the adjusted weight as a basis.
After the fault early warning model is established, the data server obtains some parameters, which may be electric energy parameters in a period of time or a certain number of electric energy parameters, then inputs the parameters into the power grid fault early warning model, and then outputs some parameter values according to the power grid fault early warning model, wherein the parameter values are predicted loads of the transformer substation, such as working parameters of a certain power utilization module of the transformer substation in a period of time or at a certain time point in the future. If the predicted load of the transformer substation is abnormal, if the numerical value is larger than the normal numerical value, the load of the transformer substation is judged to be abnormal in the time period or at the time point, and the transformer substation breaks down, so that the working state of the current transformer substation can be adjusted according to the load predicted by the power grid fault early warning model, the transformer substation is prevented from breaking down, and the use reliability is improved.
In one embodiment, the front-end sensor comprises an infrared meter reading device and a collector, the infrared meter reading device and the collector are both in wireless communication with the data server, the electric energy parameters comprise station power consumption of a power utilization module in a power station and main transformer transfer capacity, the infrared meter reading device is used for detecting the station power consumption and sending the station power consumption to the data server, and the collector is used for collecting the main transformer transfer capacity and sending the main transformer transfer capacity to the data server.
Specifically, the front-end sensors of different types are adopted to detect electric energy parameters of different types, the power consumption of the station is obtained through the infrared meter reading device, the main transformer transfer capacity is obtained through the collector, and the electric energy parameters of the demands can be obtained. The infrared meter reader follows a DL/T645-2007 protocol when acquiring data, referring to FIG. 2, when the number of the station power consumption is more than two, taking two as an example, the infrared meter reader can acquire data of the 1# station power consumption electric meter and data of the 2# station power consumption electric meter and send the data to the data server. The type of the collector is not unique, in this embodiment, the collector is an RS485 collector, taking the number of the main transformer transfer capacity data as 3 as an example, the data of the 1# main transformer transfer capacity electricity meter, the data of the 2# main transformer transfer capacity electricity meter and the data of the 3# main transformer transfer capacity electricity meter are collected through the RS485 collector, that is, are transmitted to the data server through RS485 serial port communication, so that the data server can perform further processing. It is understood that in other embodiments, the front-end sensor may comprise other types of devices and the power parameter may comprise other types of data, as deemed practicable by those skilled in the art.
In one embodiment, please refer to fig. 3, the substation power consumption analysis system further includes a centralized gateway, one side of the centralized gateway is connected to the front-end sensor, and the other side of the centralized gateway is wirelessly connected to the data server. The centralized gateway is used as a transmission medium between the front-end sensor and the data server to realize the function of data forwarding.
Specifically, the centralized gateway and the front-end sensor may be in wired communication, for example, RS485 serial port communication, and the centralized gateway and the data server are in wireless connection. The number of the centralized gateways is not unique, for example, the front-end sensor comprises an infrared meter reading device and a collector, the number of the centralized gateways can be two, including a 1# centralized gateway and a 2# centralized gateway, the data of the electricity consumption meter of the 1# station and the data of the electricity consumption meter of the 2# station are obtained through the infrared meter reading device (following a DL/T645-2007 protocol), and then the collected electricity consumption data of the stations are transmitted to the 1# centralized gateway through RS485 serial port communication. Data of the 1# main transformer transfer capacity electric meter, data of the 2# main transformer transfer capacity electric meter and data of the 3# main transformer transfer capacity electric meter are communicated through RS485 serial ports, the collected data of the main transformer transfer capacity electric meter are transmitted to the 2# centralized gateway, and the centralized gateway transmits the data to the data server through a wireless network. The main transformer transfer capacity watt-hour meter can also be detected by an infrared meter reader, and can be adjusted according to the field condition, so long as the technical personnel in the field think that the realization can be realized.
In one embodiment, the centralized gateway is wirelessly connected to the data server via a WAPI (Wireless LAN Authentication and privacy Infrastructure). The wireless connection between the WAPI and the data server can ensure the high security of the transmitted data.
Specifically, the WAPI is a security protocol and is also a mandatory security standard of the China wireless local area network. The WAPI is one of the wireless transmission protocols like infrared, Bluetooth, GPRS, CDMA1X, etc., but is different from the WAPI in that it is a transmission protocol in a Wireless Local Area Network (WLAN), and is a technology in the same field as the 802.11 transmission protocol. The application has been promised by the institute IEEE Registration Authority, authorized by the International organization for standardization ISO/IEC, to have been issued with the Ethertype field (0x88B4) assigned for the WAPI protocol. It is understood that in other embodiments, other means for wirelessly communicating with the data server may be used, such as RS485, ethernet, WIFI, LORA, 4G, NBIOT, etc., as long as those skilled in the art recognize that the implementation is possible.
According to the power utilization analysis system for the transformer substation, the front-end sensor and the data server are in communication in a wireless mode, the front-end sensor detects the electric energy parameters of the transformer substation and sends the electric energy parameters to the data server, the data server analyzes the electric energy parameters based on the BP neural network, a power grid fault early warning model is established, and the power grid fault early warning model is used for predicting the load of the transformer substation. The front-end sensor and the data server communicate in a wireless mode, the use cost can be reduced, communication lines do not need to be laid, the safety of the transformer substation is improved, the data server can introduce a BP (back propagation) neural network after acquiring electric energy parameters of the transformer substation, data analysis modeling is carried out on the electric energy parameters, a fault early warning model based on the electric energy parameters is obtained, the model can be used for predicting the load of the transformer substation, the faults are pre-judged in advance, so that solution measures can be taken in time, and the use reliability of the transformer substation power consumption analysis system is improved.
In one embodiment, please refer to fig. 4, a substation power consumption analysis method is provided, which includes the following steps:
step S200: and acquiring the electric energy parameters of the transformer substation detected by the front-end sensor.
Specifically, the data server may obtain the electric energy parameter of the substation detected by the front-end sensor. The front-end sensor detects the electric energy parameters of all the electricity utilization modules in the transformer substation and sends the electric energy parameters to the data server, the number and the types of the electricity utilization modules are not unique, generally speaking, the more the number and the types of the electricity utilization modules are, the richer the data samples collected by the front-end sensor are, and the more the power grid fault early warning model obtained based on the abundant electric energy parameters is close to the real condition, the more accurate the power grid fault early warning model is. The number of the front-end sensors is not unique, when the number of the power utilization modules is more than two, a plurality of front-end sensors can be adopted, sensors which are not used correspond to electric energy parameters of different power utilization modules, so that the detection efficiency is improved, or the same front-end sensor can be adopted to detect the electric energy parameters of the power utilization modules, so that the use cost is reduced.
Step S400: and analyzing the electric energy parameters based on the BP neural network, and establishing a power grid fault early warning model. The BP neural network is a multi-layer feedforward neural network, the model of which comprises an initial input layer, an intermediate hidden layer and a final output layer, and the algorithm of the BP neural network comprises two processes of forward propagation of signals and backward propagation of errors. When the signal is transmitted in the forward direction, the input signal acts on the output node through the middle hidden layer, and the calculated output value is generated through nonlinear transformation. And comparing the calculated output value with an expected output value (the real-time information of the electric energy parameter of the transformer substation), and generating an output error when the calculated output value is different from the expected value, so that the error is transferred to a back propagation process. The error back transmission is to back transmit the output error to the input layer through the middle hidden layer, and distribute the error to all units of each layer, and the distributed error semaphore is used as the basis for adjusting the weight of each unit. And repeatedly iterating the network parameters by using the minimum error as an approaching target through a gradient descent method, thereby determining the network parameters corresponding to the minimum error, recording the adjusted weight, and establishing a fault early warning model by using the adjusted weight as a basis.
Step S600: and predicting the load of the transformer substation according to the power grid fault early warning model.
After the fault early warning model is established, the data server obtains some parameters, which may be electric energy parameters in a period of time or a certain number of electric energy parameters, then inputs the parameters into the power grid fault early warning model, and then outputs some parameter values according to the power grid fault early warning model, wherein the parameter values are predicted loads of the transformer substation, such as working parameters of a certain power utilization module of the transformer substation in a period of time or at a certain time point in the future. If the predicted load of the transformer substation is abnormal, if the numerical value is larger than the normal numerical value, the load of the transformer substation is judged to be abnormal in the time period or at the time point, and the transformer substation breaks down, so that the working state of the current transformer substation can be adjusted according to the load predicted by the power grid fault early warning model, the transformer substation is prevented from breaking down, and the use reliability is improved.
In one embodiment, the front-end sensor includes an infrared meter reader and a collector, and the electric energy parameters include station power consumption and main transformer transfer capacity of the power utilization module in the power station, see fig. 5, and step S200 includes step S220.
Step S220: and acquiring the station power consumption detected by the infrared meter reader and the main transformer transfer capacity acquired by the collector.
Specifically, the front-end sensors of different types are adopted to detect electric energy parameters of different types, the power consumption of the station is obtained through the infrared meter reading device, the main transformer transfer capacity is obtained through the collector, and the electric energy parameters of the demands can be obtained. The infrared meter reader follows DL/T645-2007 protocol when acquiring data, and when the number of the station electricity consumption is more than two, taking two as an example, the infrared meter reader can acquire data of the 1# station electricity consumption electric meter and data of the 2# station electricity consumption electric meter and send the data to the data server. The type of the collector is not unique, in this embodiment, the collector is an RS485 collector, taking the number of the main transformer transfer capacity data as 3 as an example, the data of the 1# main transformer transfer capacity electricity meter, the data of the 2# main transformer transfer capacity electricity meter and the data of the 3# main transformer transfer capacity electricity meter are collected through the RS485 collector, that is, are transmitted to the data server through RS485 serial port communication, so that the data server can perform further processing. It is understood that in other embodiments, the front-end sensor may comprise other types of devices and the power parameter may comprise other types of data, as deemed practicable by those skilled in the art.
In one embodiment, referring to fig. 5, step S400 includes steps S420 to S460.
Step S420: and calculating according to the station power consumption and the main transformer transfer capacity to obtain the station power consumption.
Specifically, the station power consumption can be calculated by a formula "station power consumption is station power consumption/main transformer transfer capacity", and further, when the number of the station power consumption and the main transformer transfer capacity is more than two, the station power consumption in the formula is the sum of the plurality of station power consumptions, and the main transformer transfer capacity in the formula is the sum of all the main transformer transfer capacities.
Step S440: and calculating the station electricity utilization change rate according to the station electricity utilization rate and the preset station electricity utilization rate.
The type of the preset station electricity consumption rate is not unique and can be adjusted according to the type of the station electricity consumption change rate required. In this embodiment, taking the preset station power consumption rate as the power consumption rate of the station in the same period of the last year as an example, and the station power consumption change rate is the rate of increase or decrease of the station power consumption rate in the same ratio (the same period of the last year), the preset station power consumption rate may be stored in the data server in advance and may be called directly when needed, and generally, the data storage time is not less than 10 years. It is understood that the preset station power usage may be of other types in other embodiments. For example, the power consumption of the stations in the same period of the last month is adjusted according to actual requirements.
Step S460: and analyzing the station power consumption rate and the station power consumption change rate based on the BP neural network, and establishing a power grid fault early warning model.
After the power consumption rate of the arriving station and the power consumption change rate of the station are obtained, the background of the data server starts to calculate and compare, and the following results can be obtained: the method comprises the steps of comparing station power consumption and station power consumption rates of different areas at the same time, comparing station power consumption and station power consumption rates of the same area at different times, comparing station power consumption and station power consumption rates of different devices of the same type at the same temperature and comparing station power consumption and station power consumption rates of the same devices at different temperatures, wherein data can be used as input information sources of a BP neural network, the BP neural network generates output values through BP neural network operation according to collected information quantities of a plurality of sources as input signals, and fault early warning models are sequentially established.
In one embodiment, referring to fig. 6, step S460 includes steps S462 to S466.
Step S462: and taking the power consumption rate and the power consumption change rate of the station as input information sources of the BP neural network.
Referring to fig. 10, the BP neural network generates an output value through operation of the BP neural network according to the collected information amount from a plurality of sources as an input signal. In the prediction of the power consumption, in order to comprehensively analyze the problem, the defects that the calculation speed of a BP neural network is relatively slow and the BP neural network is easy to sink to a local extreme point are overcome, so that the prediction result is more accurate, a plurality of related variables are provided, and certain correlation exists among the variables, and the correlation influences the prediction accuracy of the power consumption. In this embodiment, the station power consumption rate and the station power consumption change rate are used as input information sources of the BP neural network, and in order to improve the accuracy, the types of the input information sources may be expanded, such as a STATCOM water cooling system information acquisition unit, a main transformer fan information acquisition unit, a station power consumption feeder line information acquisition unit, a protection room information acquisition unit, and a main transformer, and related quantities such as transmission power, environmental temperature and humidity. It is understood that in other embodiments, other input information sources, such as voltage, etc., may be introduced, and may be adjusted according to actual requirements.
Step S464: and acquiring the node number of the hidden layer of the BP neural network.
The number of nodes of the hidden layer is an important parameter affecting the performance of the BP neural network, and generally, the greater the number of nodes of the hidden layer, the better the performance of the BP network. The range of the number of nodes of the hidden layer of the neural network can be calculated to be 7-13 according to an empirical formula, and the number of the nodes of the hidden layer can be set to be 13 maximum neuron nodes in the embodiment.
Step S466: and training the BP neural network according to the LM algorithm based on the input information source and the node number, and establishing a power grid fault early warning model.
Because the BP neural network is easy to fall into local optimum, in order to avoid the situation, an LM algorithm is selected as a training method to train the BP neural network. The BP neural network takes the minimum error as an approximate target, and repeatedly iterates network parameters by a gradient descent method so as to determine the network parameters corresponding to the minimum error, namely E (preset precision), and record the weight value adjusted at the moment.
In the formula: (z)i-di)2Is the absolute error of the desired output from the network from the actual output; n is the number of learning samples.
Calculating an implicit node number formula:
wherein k is the sample size; n is1For implicit tier node numbers, n is the input tier node number. When i is>n1When the temperature of the water is higher than the set temperature,
n1＝(n+m)1/2+a (3)
wherein n is the number of nodes of the input layer; m is the number of output layer nodes and a is a constant between 1 and 10.
n1＝log2n (4)
And (4) simultaneous 2-4 formula solution, and the number interval [ a, b ] of the nodes of the hidden layer can be obtained.
The LM algorithm formula is:
xs+1＝xs-(H+αI)G (5)
where H is the Hessian matrix of the multidimensional vector, G is the first order gradient of the multidimensional vector, α is the step size, and I is the identity matrix. When the descending is too fast, a smaller alpha is used, so that the whole formula is close to the Gauss-Newton method, and when the descending is too slow, a larger alpha is used, so that the whole formula is close to the gradient method, thereby establishing a power grid fault early warning model.
Referring to fig. 11, the LM algorithm is used as a training method to perform BP neural network training on the station power consumption load training samples, and a power grid fault early warning model is established. In the figure, X is the input sample set, Y is the desired output, G is the net output, and E outputs the error. And (3) creating a three-layer BP neural network by using a BP neural network toolbox in MATLAB, setting the transfer function of the hidden layer as logsin, setting the transfer function of the output layer as purelin, and setting the training function as train, the learning rate is 0.2 and the target precision is 0.005. It is understood that in other embodiments, other algorithms may be used to train the BP neural network, as long as those skilled in the art recognize that the training can be achieved.
In one embodiment, referring to fig. 6, step S420 includes steps S422 to S426.
Step S422: and acquiring historical maximum power consumption corresponding to the station power consumption in a preset acquisition period.
The specific value of the preset acquisition period is not unique and can be adjusted according to actual requirements. The collected station power consumption is time sequence data, i.e. the time distribution of data, such as what data is 5 o 'clock and what data is half 5 o' clock, after the data are collected, the maximum power consumption is selected from the data, and further, for further analysis, the time when the historical maximum power consumption appears can be correspondingly stored.
Step S424: and taking the historical maximum power consumption as a reference, and carrying out normalization processing on the station power consumption.
The learning efficiency of the BP neural network can be improved by normalizing the station power consumption, and the normalization of each input information source of the BP neural network can be realized in an extensible manner, so that the power consumption normalization of the STATCOM system and the power consumption normalization of the transformer are taken as examples, and the power consumption normalization of the STATCOM system is realized: the method is characterized in that the power consumption data is normalized according to the limit power consumption of the water cooling system counted for many years:
in the formula: whThe normalized STATCOM system power consumption value is obtained; wtThe real-time electricity consumption value is acquired by the STATCOM electricity quantity acquisition unit; wmaxThe method is used for monitoring the historical maximum power consumption of the STATCOM power acquisition unit. And (3) normalizing the electric quantity of the transformer: normalizing the power consumption data into power consumption data according to the limit power consumption of the water cooling system counted for many years
In the formula: whThe normalized transformer electric quantity value is obtained; wtThe real-time power consumption value is acquired by the transformer power acquisition unit; wmaxThe historical maximum power consumption monitored by the transformer power acquisition unit is obtained. The normalization of the electric quantity of the alternating current feeder, the electric quantity of the protection room and the humidity of the protection room adopts a method similar to the above normalization. The data are normalized, so that the learning efficiency of the BP neural network can be improved, the neuron saturation is avoided, and the prediction precision is improved.
Step S426: and calculating the station power consumption according to the station power consumption and the main transformer transfer capacity after the normalization treatment.
The power consumption of the station is calculated according to the power consumption of the station after normalization processing, so that when the power consumption of the station is used as one of input information sources of the BP neural network, the learning efficiency of the BP neural network is improved, neuron saturation is avoided, and the prediction precision is improved.
In one embodiment, referring to fig. 7, step S200 includes step S202 and step S204.
Step S202: and performing network security authentication on the electric energy parameters of the transformer substation detected by the front-end sensor.
Specifically, the type of network security authentication is not unique, and in this embodiment, please refer to fig. 8, 802.1x network security authentication and WAPI wireless network security authentication are performed on the power parameters of the substation detected by the front-end sensor, and the authenticated data can be received by the data server, so that the security of data transmission is improved. The 802.1X protocol is a port-based network access control protocol, and therefore a specific 802.1X authentication function must be configured on the device port. A user device accessing on a port must be authenticated to control access to network resources. The 802.1X authentication system employs a typical network application system client/server (C/S) architecture, including sequentially connected client, device, and server 3 portions. Client side: the LAN user terminal device, but it must be EAPOL (extensible authentication protocol over LAN) to support the device (e.g., PC), may initiate 802.1X authentication by launching 802.1X client software installed on the client device. The Device end: a network device (e.g., switch) supporting the 802.1X protocol performs an identity check on a connected client. The Server authentication Server: the device providing the authentication service for the 802.1X protocol at the device end is the true authentication, the working diagram of the WAPI authentication is shown in fig. 9, 802.1X is a basic protocol, the international universal standard, the WAPI is a safer communication standard for the field of power metering, and the security of the two combination modes is higher.
When the data server is directly communicated with the front-end sensor, the data server can carry out network security authentication on the electric energy parameters of the transformer substation detected by the front-end sensor, and when the data server is communicated with the front-end sensor through the centralized gateway, the centralized gateway can carry out network security authentication on the electric energy parameters of the transformer substation detected by the front-end sensor. Further, when the electric energy parameter comprises various types of data, network security authentication can be performed on the various types of data so as to improve the security of data transmission.
Step S204: and receiving the electric energy parameters after the electric energy parameters pass the network security authentication.
Network security authentication is carried out on the electric energy parameters of the transformer substation from the front-end sensor, and the authenticated data can be received by the data server, so that the security of data transmission is improved.
For a better understanding of the above embodiments, the following detailed description is given in conjunction with a specific embodiment. In one embodiment, the substation electricity utilization analysis method comprises the following steps: the data of the electricity consumption meter of the 1# station and the data of the electricity consumption meter of the 2# station are obtained through an infrared meter reader (following a DL/T645-2007 protocol), and then the collected electricity consumption data of the stations are transmitted to the 1# centralized gateway through RS485 serial port communication. The data of the 1# main transformer transfer capacity electric meter, the data of the 2# main transformer transfer capacity electric meter and the data of the 3# main transformer transfer capacity electric meter are communicated through RS485 serial ports, the collected data of the main transformer transfer capacity electric meter are transmitted to the 2# centralized gateway, and the data are transmitted to the data server through the centralized gateway and the WAPI wireless network. Calculating the outbound power usage by the following formula: the power consumption rate of the station is equal to the power consumption of the station/the main transformer transfer capacity, and the power consumption change rate of the station is calculated by the following formula: the station power utilization rate is the rate of increase or decrease of the station power utilization rate in the same ratio (in the same period as the last year).
According to the formula, the background computer starts to calculate and compare, and the following four comparison results are obtained respectively: the method comprises the steps of comparing station power consumption and station power consumption rate in different areas at the same time, comparing station power consumption and station power consumption rate in the same area at different times, comparing station power consumption and station power consumption rate of the same type but different equipment at the same temperature and comparing station power consumption and station power consumption rate of the same equipment at different temperatures.
The centralized gateway respectively carries out 802.1x network security authentication on the data of the 1# centralized gateway and the 2# centralized gateway and carries out WAPI wireless network security authentication through the WAPI wireless network. The authenticated data can be transmitted to the data server. The 802.1X protocol is a port-based network access control protocol, and therefore a specific 802.1X authentication function must be configured on the device port. A user device accessing on a port must be authenticated to control access to network resources. The 802.1X authentication system employs a typical network application system client/server (C/S) architecture, including client, device, and server 3 portions. Client side: the LAN user terminal device, but it must be EAPOL (extensible authentication protocol over LAN) to support the device (e.g., PC), may initiate 802.1X authentication by launching 802.1X client software installed on the client device. The Device end: a network device (e.g., switch) supporting the 802.1X protocol performs an identity check on a connected client. The Server authentication Server: the device that provides authentication services for the 802.1X protocol at the device side is a true authentication.
The WAPI is a safety protocol and is also a mandatory safety standard of the China wireless local area network. The WAPI is one of the wireless transmission protocols like infrared, Bluetooth, GPRS, CDMA1X, etc., but is different from the WAPI in that it is a transmission protocol in a Wireless Local Area Network (WLAN), and is a technology in the same field as the 802.11 transmission protocol. The scheme has been promised by the institute IEEE Registration Authority authorized by the international organization for standardization ISO/IEC, and has been assigned the ethertype field (0x88B4) for the WAPI protocol.
And optimizing initial weight of the BP neural network by comparison and introducing an improved genetic algorithm to carry out data mining comparison. The BP neural network is a multi-layer feedforward neural network, the model of which comprises an initial input layer, an intermediate hidden layer and a final output layer, and the algorithm of the BP neural network comprises two processes of forward propagation of signals and backward propagation of errors. When the signal is transmitted in the forward direction, the input signal acts on the output node through the middle hidden layer, and the calculated output value is generated through nonlinear transformation. Comparing the calculated output value with the expected output value (real-time information of the system herein), when the calculated output value is different from the expected value, an output error will be generated, and then the process of back propagation of the error will be carried out. The error back transmission is to back transmit the output error to the input layer through the middle hidden layer, and distribute the error to all units of each layer, and the distributed error semaphore is used as the basis for adjusting the weight of each unit. And (3) with the minimum error as an approaching target, repeatedly iterating the network parameters by a gradient descent method so as to determine the network parameter corresponding to the minimum error, namely E (preset precision), and recording the adjusted weight.
In the formula: (z)i-di)2Is the absolute error of the desired output from the network from the actual output; n is the number of learning samples.
The BP neural network with multiple information sources is used for generating an output value through calculation of the BP neural network according to the collected information quantity of the multiple sources as an input signal. In the prediction of the power consumption, in order to comprehensively analyze the problem, the defects that the calculation speed of a BP neural network is relatively slow and the BP neural network is easy to sink to a local extreme point are overcome, so that the prediction result is more accurate, a plurality of related variables are provided, and certain correlation exists among the variables, and the correlation influences the prediction accuracy of the power consumption. The BP neural network model adopts a STATCOM water cooling system information acquisition unit, a main transformer fan information acquisition unit, a station power utilization feeder line information acquisition unit, a protection room information acquisition unit, main transformer transmission electric quantity, environment temperature and humidity and other related quantities as input information sources.
The range of the number of the nodes of the hidden layer of the neural network can be calculated to be 7-13 according to an empirical formula, and the number of the nodes of the hidden layer has great influence on the performance of the BP neural network. In general, the greater the number of nodes of the hidden layer, the better the performance of the BP network, and therefore, the number of nodes of the hidden layer is set to the maximum 13 neuron nodes herein. Since the BP neural network is easy to fall into local optimum, in order to avoid the situation, the LM algorithm is selected as a training method to train the BP network.
Calculating an implicit node number formula:
wherein k is the sample size; n1 is the number of hidden layer nodes and n is the number of input layer nodes. When i is>n1When the temperature of the water is higher than the set temperature,
n1＝(n+m)1/2+a (3)
wherein n is the number of nodes of the input layer; m is the number of output layer nodes and a is a constant between 1 and 10.
n1＝log2n (4)
And (4) simultaneous 2-4 formula solution, and the number interval [ a, b ] of the nodes of the hidden layer can be obtained.
The LM algorithm formula is:
xs+1＝xs-(H+αI)G (5)
where H is the Hessian matrix of the multidimensional vector, G is the first order gradient of the multidimensional vector, α is the step size, and I is the identity matrix. When the drop is too fast, a smaller alpha is used, so that the whole formula is close to the Gauss-Newton method; when the fall is too slow, a larger α is used, making the whole formula close to the gradient method.
Factors influencing the load of the station electric equipment mainly comprise a STATCOM water cooling system unit, a main transformer fan unit, a station electric feeder unit, a protection room unit, and a main transformer electric transmission quantity and environment temperature and humidity. These data constitute the basic input data space of a BP neural network with multiple information sources, and the output of the neural network is the electricity consumption of the station electric equipment. The input quantities were collected as follows:
the STATCOM electric quantity acquisition unit is used for acquiring real-time quantity (water supply flow temperature and pressure, water return temperature, air inlet temperature, pressure of an inlet of a main circulating pump and a buffer tank, conductivity of cooling water and deionized water, liquid level of the buffer tank, and temperature and humidity of a valve hall); the main circulating pump, the electric heater, the fluid infusion pump, the air supply electromagnetic valve and the fan.
The transformer electric quantity acquisition unit: the main electric quantity, the oil temperature, the variable-current loop temperature, the electric weight of the fan and the rotating speed are changed.
The alternating current feeder line electric quantity acquisition unit: general circuit, power, illumination, direct current machine that charges, office protection room environment and electric quantity acquisition unit: temperature and humidity, air conditioner electric quantity and illumination electric quantity.
In order to improve the learning efficiency of a BP neural network with multiple information sources, avoid neuron saturation and improve prediction precision, the data is normalized: the STATCOM system is normalized by power consumption. The method is characterized in that the power consumption data is normalized according to the limit power consumption of the water cooling system counted for many years:
in the formula: whThe normalized STATCOM system power consumption value is obtained; wtThe real-time electricity consumption value is acquired by the STATCOM electricity quantity acquisition unit; wmaxThe method is used for monitoring the historical maximum power consumption of the STATCOM power acquisition unit.
And normalizing the electric quantity of the transformer. Normalizing the power consumption data into power consumption data according to the limit power consumption of the water cooling system counted for many years
In the formula: whThe normalized transformer electric quantity value is obtained; wtThe real-time power consumption value is acquired by the transformer power acquisition unit; wmaxThe historical maximum power consumption monitored by the transformer power acquisition unit is obtained. The normalization of the electric quantity of the alternating current feeder, the electric quantity of the protection room and the humidity of the protection room adopts a method similar to the above normalization.
The process of establishing the power grid fault early warning model can be simply summarized as follows:
(1) collecting time sequence data of electric loads of 4 monitoring points (a water cooling system, a main transformer fan, a station electric feeder line and a protection room);
(2) selecting historical maximum power consumption monitored by the power acquisition unit of each monitoring point as a reference, and carrying out normalization processing on data;
(3) calculating the node number range of the hidden layer of the neural network by an empirical formula, and determining the node number range to be 13;
(4) and (3) performing BP neural network training on the station power consumption load training sample by using an LM algorithm as a training method, and establishing a cloud computing resource load prediction model.
The predictive model is shown in fig. 6. In the figure, X is the input sample set, Y is the desired output, G is the net output, and E outputs the error. And (3) creating a three-layer BP neural network by using a BP neural network toolbox in MATLAB, setting the transfer function of the hidden layer as logsin, setting the transfer function of the output layer as purelin, and setting the training function as train, the learning rate is 0.2 and the target precision is 0.005.
The method and the system provided by the application are adopted to predict the load of the electrical equipment of a certain 500kV transformer substation in Guangdong province. And (3) carrying out predictive analysis on 672 station power consumption time series data from 3/1/2018 to 3/7/2018, taking the first 500 data as sample data, finishing the training of the multi-information-source BP neural network prediction model, predicting the last 172 data, and finishing the verification of the prediction network model. Analysis shows that the multi-information-source BP neural network can effectively track the change situation of the substation electric equipment load, accurately describe the next substation electric equipment load and obtain a high-precision substation electric equipment load prediction result.
In order to test the superiority of the station electric equipment load prediction of the multi-information-source BP neural network, three schemes of a regression analysis method, a grey system theory and an index smoothing method are selected to perform station electric load prediction respectively. And comparing the accuracy of the prediction results of the four schemes, finding that the prediction accuracy of the model is the highest, and better overcoming the defects of the load prediction models of the electric equipment of other stations.
According to the transformer substation power consumption analysis method, the front-end sensor and the data server are in wireless communication, the front-end sensor detects the electric energy parameters of the transformer substation and sends the electric energy parameters to the data server, the data server analyzes the electric energy parameters based on the BP neural network, a power grid fault early warning model is established, and the power grid fault early warning model is used for predicting the load of the transformer substation. The front-end sensor and the data server communicate in a wireless mode, the use cost can be reduced, communication lines do not need to be laid, the safety of the transformer substation is improved, the data server can introduce a BP (back propagation) neural network after acquiring electric energy parameters of the transformer substation, data analysis modeling is carried out on the electric energy parameters, a fault early warning model based on the electric energy parameters is obtained, the model can be used for predicting the load of the transformer substation, the faults are pre-judged in advance, so that solution measures can be taken in time, and the use reliability of the transformer substation power consumption analysis system is improved.
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. The transformer substation electricity utilization analysis system is characterized by comprising a front-end sensor and a data server, wherein the front-end sensor and the data server are communicated in a wireless mode;
the front-end sensor is used for detecting electric energy parameters of the transformer substation and sending the electric energy parameters to the data server, and the data server is used for analyzing the electric energy parameters based on a BP neural network, establishing a power grid fault early warning model and predicting the load of the transformer substation according to the power grid fault early warning model.
2. The substation power consumption analysis system according to claim 1, wherein the front-end sensor comprises an infrared meter reader and a collector, the infrared meter reader and the collector are both in wireless communication with the data server, and the electric energy parameters comprise station power consumption and main transformer transfer capacity of a power utilization module in a power utilization station;
the infrared meter reading device is used for detecting the station power consumption and sending the station power consumption to the data server, and the collector is used for collecting the main transformer transfer capacity and sending the main transformer transfer capacity to the data server.
3. The substation power consumption analysis system of claim 1, further comprising a centralized gateway, wherein one side of the centralized gateway is connected to the front-end sensor, and the other side of the centralized gateway is wirelessly connected to the data server.
4. The substation power consumption analysis system of claim 3, wherein the centralized gateway is wirelessly connected to the data server via a WAPI wireless network.
5. A transformer substation electricity utilization analysis method is characterized by comprising the following steps:
acquiring electric energy parameters of a transformer substation detected by a front-end sensor;
analyzing the electric energy parameters based on a BP neural network, and establishing a power grid fault early warning model;
and predicting the load of the transformer substation according to the power grid fault early warning model.
6. The substation power consumption analysis method according to claim 5, wherein the front-end sensor comprises an infrared meter reader and a collector, the electric energy parameters comprise station power consumption and main transformer transfer capacity of a power consumption module in a power station, and the acquiring of the electric energy parameters of the substation detected by the front-end sensor comprises:
and acquiring the station power consumption detected by the infrared meter reader and the main transformer transfer capacity acquired by the collector.
7. The substation power consumption analysis method according to claim 6, wherein the analyzing the electric energy parameters based on the BP neural network and establishing a power grid fault early warning model comprises:
calculating to obtain station power consumption according to the station power consumption and the main transformer transfer capacity;
calculating to obtain a station electricity utilization change rate according to the station electricity utilization rate and a preset station electricity utilization rate;
and analyzing the station power consumption rate and the station power consumption change rate based on the BP neural network, and establishing a power grid fault early warning model.
8. The substation power consumption analysis method according to claim 7, wherein the analyzing the station power consumption rate and the station power consumption change rate based on the BP neural network, and the establishing of the power grid fault early warning model comprises:
taking the station electricity utilization rate and the station electricity utilization change rate as input information sources of the BP neural network;
acquiring the number of nodes of a hidden layer of the BP neural network;
and training the BP neural network according to an LM algorithm based on the input information source and the node number, and establishing a power grid fault early warning model.
9. The substation power analysis method according to claim 7, wherein the calculating the substation power consumption according to the substation power consumption and the main transformer transfer capacity comprises:
acquiring historical maximum power consumption corresponding to the station power consumption in a preset acquisition period;
taking the historical maximum power consumption as a reference, and carrying out normalization processing on the station power consumption;
and calculating to obtain the station power consumption according to the station power consumption after the normalization processing and the main transformer transfer capacity.
10. The substation power consumption analysis method according to claim 5, wherein the obtaining of the electric energy parameter of the substation detected by the front-end sensor comprises:
network security authentication is carried out on the electric energy parameters of the transformer substation detected by the front-end sensor;
and receiving the electric energy parameter after the electric energy parameter passes the network security authentication.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010450628.7A CN111680831A (en) | 2020-05-25 | 2020-05-25 | Substation power utilization analysis system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010450628.7A CN111680831A (en) | 2020-05-25 | 2020-05-25 | Substation power utilization analysis system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111680831A true CN111680831A (en) | 2020-09-18 |
Family
ID=72434363
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010450628.7A Pending CN111680831A (en) | 2020-05-25 | 2020-05-25 | Substation power utilization analysis system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111680831A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112184487A (en) * | 2020-09-30 | 2021-01-05 | 国网北京市电力公司 | Method and device for predicting power supply index |
| CN113420510A (en) * | 2021-07-07 | 2021-09-21 | 广东电网有限责任公司 | Energy consumption distinguishing method based on front-end sensing and learning |
| CN113514739A (en) * | 2021-06-16 | 2021-10-19 | 国网吉林省电力有限公司电力科学研究院 | IWOA-BP algorithm-based oil paper insulation aging evaluation method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101771275A (en) * | 2008-12-31 | 2010-07-07 | 中国神华能源股份有限公司 | Electric energy monitoring system |
| US20110251732A1 (en) * | 2010-04-10 | 2011-10-13 | Schweitzer Iii Edmund O | Systems and method for obtaining a load model and related parameters based on load dynamics |
| CN104809337A (en) * | 2015-04-15 | 2015-07-29 | 华南理工大学 | Practical model construction method for electricity consumption of main station loads of transformer substation |
| CN104992246A (en) * | 2015-07-09 | 2015-10-21 | 华南理工大学 | Improved-least-square-method-based prediction method of load electric quantity for transformer substation |
| CN109102171A (en) * | 2018-07-24 | 2018-12-28 | 上海欣影电力科技股份有限公司 | A kind of substation equipment condition intelligent evaluation system and method based on big data |
| CN110263839A (en) * | 2019-06-13 | 2019-09-20 | 河海大学 | Power system load static characteristic online intelligent recognition method based on big data |
| CN110994798A (en) * | 2019-12-16 | 2020-04-10 | 深圳供电局有限公司 | Substation equipment monitoring system |
| CN111144638A (en) * | 2019-12-24 | 2020-05-12 | 东南大学 | Power distribution network operation situation prediction method based on big data |
-
2020
- 2020-05-25 CN CN202010450628.7A patent/CN111680831A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101771275A (en) * | 2008-12-31 | 2010-07-07 | 中国神华能源股份有限公司 | Electric energy monitoring system |
| US20110251732A1 (en) * | 2010-04-10 | 2011-10-13 | Schweitzer Iii Edmund O | Systems and method for obtaining a load model and related parameters based on load dynamics |
| CN104809337A (en) * | 2015-04-15 | 2015-07-29 | 华南理工大学 | Practical model construction method for electricity consumption of main station loads of transformer substation |
| CN104992246A (en) * | 2015-07-09 | 2015-10-21 | 华南理工大学 | Improved-least-square-method-based prediction method of load electric quantity for transformer substation |
| CN109102171A (en) * | 2018-07-24 | 2018-12-28 | 上海欣影电力科技股份有限公司 | A kind of substation equipment condition intelligent evaluation system and method based on big data |
| CN110263839A (en) * | 2019-06-13 | 2019-09-20 | 河海大学 | Power system load static characteristic online intelligent recognition method based on big data |
| CN110994798A (en) * | 2019-12-16 | 2020-04-10 | 深圳供电局有限公司 | Substation equipment monitoring system |
| CN111144638A (en) * | 2019-12-24 | 2020-05-12 | 东南大学 | Power distribution network operation situation prediction method based on big data |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112184487A (en) * | 2020-09-30 | 2021-01-05 | 国网北京市电力公司 | Method and device for predicting power supply index |
| CN113514739A (en) * | 2021-06-16 | 2021-10-19 | 国网吉林省电力有限公司电力科学研究院 | IWOA-BP algorithm-based oil paper insulation aging evaluation method |
| CN113420510A (en) * | 2021-07-07 | 2021-09-21 | 广东电网有限责任公司 | Energy consumption distinguishing method based on front-end sensing and learning |
| CN113420510B (en) * | 2021-07-07 | 2022-06-17 | 广东电网有限责任公司 | Energy consumption distinguishing method based on front-end sensing and learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111680831A (en) | Substation power utilization analysis system and method | |
| Zhao et al. | Development of an energy monitoring system for large public buildings | |
| CN104570973B (en) | Method and device for enterprise energy management | |
| CN105804981B (en) | Motor train unit traction transformer oil pump remote failure monitors and early warning system and its method | |
| US20110153107A1 (en) | Apparatus and method for smart energy management by controlling power consumption | |
| CN111625991A (en) | Low-voltage distribution network topology verification method | |
| CN110910626A (en) | Energy consumption information acquisition system based on ubiquitous power Internet of things | |
| CN108508396A (en) | Multiple-in-one collecting device detecting system and method | |
| CN105093047B (en) | Power circuit load monitoring early warning system | |
| CN110018371B (en) | Converter station main equipment state maintenance system and implementation method thereof | |
| CN107463732A (en) | A kind of multiterminal alternating current-direct current active distribution network scheduling controlling analogue system and method | |
| CN107657339A (en) | Family's energy efficiency managing method and system | |
| Manoharan et al. | An Intellectual Energy Device for Household Appliances Using Artificial Neural Network | |
| CN106249628B (en) | A kind of smart machine installation site automatic recognition system and method | |
| CN211667970U (en) | Power supply diagnosis and alarm device based on air conditioning energy control of communication machine room | |
| CN107611940A (en) | A kind of power distribution network method for monitoring abnormality and system based on historical data analysis | |
| CN116317096B (en) | Drawer type sampling control cabinet system based on artificial intelligence | |
| CN101231523B (en) | Control system and method for sensibility charge | |
| CN106331162B (en) | Pipe gallery main controller based on Internet and monitoring method thereof | |
| CN109557362A (en) | A kind of anti-electricity-theft monitoring method of distribution line and system | |
| Nichiforov et al. | Embedded on-line system for electrical energy measurement and forecasting in buildings | |
| Moayedi et al. | Energy Optimization Technologies in Smart Homes | |
| CN112233402A (en) | Photovoltaic power generation on-line monitoring system and monitoring method | |
| Hu | Research on Battery Monitoring Technology Based on Internet of Things | |
| Liu et al. | [Retracted] Design and Implementation of Large‐Scale Public Building Energy Consumption Monitoring Platform Based on BP Neural Network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |