Electric power system edge calculation and data distribution device based on industrial personal computer
Technical Field
The invention belongs to the technical field of power system application terminals, and particularly relates to an edge calculation and data distribution device of a power system based on an industrial personal computer.
Background
The traditional power load monitoring adopts an intrusive method, namely, sensors are arranged on all electric equipment of users to record the use conditions of the electric equipment. The method has the advantages of accurate and reliable monitoring data and the disadvantages of poor operability, high implementation cost and low user acceptance degree. Therefore, a non-invasive load monitoring technology comes to mind, and briefly, the non-invasive load monitoring technology decomposes the total load information of a user into information of each electric device through each parameter detected by a user electric meter, and further obtains the energy consumption condition of the electric device, the electricity utilization rule of the user and other electricity utilization information.
Disclosure of Invention
The invention aims to develop a power system edge calculation and data distribution device which is strong in operability, easy to implement, accurate and reliable in detection data and capable of monitoring and feeding back the power utilization condition of a user in real time by using a non-invasive load monitoring technology on the basis of an industrial touch control integrated machine.
The technical scheme provided by the invention is as follows:
an edge calculation and data distribution device of a power system based on an industrial personal computer is characterized by comprising a core industrial control module, and an input/output module, a data acquisition module, a data processing module, a state monitoring module and a data distribution module which are respectively connected with the core industrial control module;
the input/output module is realized by adopting a touch screen and is used for receiving a user operation command, transmitting the user operation command to the core industrial control module and displaying related information according to an instruction sent by the core industrial control module;
the data acquisition module is used for carrying out information interaction with the intelligent electric meter through the interface circuit based on a control instruction of the core industrial control module, and acquiring and storing total power load data sent by the intelligent electric meter;
the core industrial control module comprises a CPU microprocessor and a memory, the CPU microprocessor is used for processing total power load data sent by the data acquisition module, respectively transmitting corresponding data in the total power load data to the state monitoring module and the data processing module, receiving and storing processing result data fed back by the two modules, and controlling the touch screen to display related information according to the processing results fed back by the two modules; the memory is used for storing the total power load data and processing result data fed back by the data processing module and the state monitoring module to construct a system database;
the data processing module is used for establishing normalized multidimensional load data for the received power load data, realizing the classification and identification of the household power load by adopting a genetic algorithm-based optimized BP neural network algorithm (GA-BP for short) based on the multidimensional load data, and feeding back an identification result to the core industrial control module;
the state monitoring module is used for comparing threshold values of the received power load data, feeding back a comparison result to the core industrial control module, and triggering an alarm signal if the related power load data exceeds the threshold value of a normal state;
the data distribution module is connected with the electric power data cloud end, and according to the setting of a user on the sending content, data are extracted from the system database and uploaded to the cloud end, and a large electric power load data set is established.
On the basis of the above scheme, a further improved or preferred scheme further comprises:
further, the data acquisition module includes a serial data sending unit, a serial data receiving unit, a data protocol analyzing unit and a data storage unit, wherein:
the serial port data sending unit sends a data packet to the intelligent electric meter according to a communication protocol between the serial port data sending unit and the intelligent electric meter, data items are represented by using a compressed BCD code, low bytes are in front, high bytes are in back, and 33H is added to each byte in each frame of data field for encryption;
receiving return data from the intelligent ammeter by a serial port data receiving unit;
the data protocol analysis unit analyzes the returned data according to a communication protocol, subtracts 33H from the data field data of the received electric meter returned frame for decryption, and extracts total power load data;
and the data storage unit stores the total power load data for the core industrial control module to fetch.
Further, the data processing module comprises a data normalization unit, a power load identification unit and an identification result display unit, wherein:
the data normalization unit is used for carrying out normalization processing on the received power load data to obtain a normalized power load data set;
identifying the normalized power load data set by a power load identification unit, optimizing weight and bias on the basis of a BP (back propagation) neural network by utilizing a genetic algorithm, constructing the BP neural network, and finally establishing a BP neural network model based on genetic algorithm optimization, namely a GA-BP network model, so as to realize classification identification of the power load and determine the type and name of the power load;
the identification result display unit is responsible for feeding back the identification result to the core industrial control module and triggering the core industrial control module to control the touch screen to display the identification result in real time.
Further, the state monitoring module comprises a state threshold setting unit, a state monitoring unit and an alarm indicating unit, wherein:
the state threshold value setting unit completes setting of the threshold value range of the load parameter in the normal state according to the instruction input by the user;
the state monitoring unit compares the received power load data with the load parameter threshold range, and sends a signal to the alarm indicating unit when the load parameter threshold range exceeds the normal state;
and the alarm indication unit feeds back a processing result to the core industrial control module, and controls the touch screen to carry out alarm indication on the load parameter exceeding the normal threshold range.
Furthermore, the core industrial control module comprises a CPU microprocessor, an RAM memory, a memory and a second bus, the second bus is used as a common channel for transmitting information inside the core industrial control module, and the RAM memory, the memory and the I/O interface are in communication connection with the CPU microprocessor through the second bus.
Preferably, a communication protocol between the serial port data sending unit and the intelligent electric meter is DL/T645-2007;
preferably, the input/output module adopts a 17-inch capacitance touch type LED liquid crystal screen SG 17A;
the type of the RAM memory is 2G/DDR 3;
the memory is a 32G solid state disk;
the CPU microprocessor is a J1800 dual-core processor.
Furthermore, the device is provided with a first bus used as a common channel for transmitting information in the device, and the data processing module, the state monitoring module and the data distribution module are in communication connection with the core industrial control module through the first bus.
Has the advantages that:
the electric power system edge calculation and data distribution device based on the industrial personal computer acquires electric power load data by using the intelligent electric meter, realizes the classification and identification of household electric power loads by adopting a BP neural network algorithm based on genetic algorithm optimization, has better global search capability, can obtain global optimal solution at a higher convergence speed, acquires user electricity utilization information by edge calculation processing, distributes the processed data to an electric power data cloud, can realize the construction of a large data set of the electric power loads, provides reliable reference data for the operation and maintenance of the electric power system, and has the advantages of high safety, low economic investment, strong practicability and the like.
Drawings
FIG. 1 is a system block diagram of the apparatus of the present invention;
FIG. 2 is a block diagram of the core industrial control module;
FIG. 3 is a flowchart of a BP neural network algorithm based on genetic algorithm optimization;
FIG. 4 is a graph of the relationship between algorithm identification accuracy and iteration number;
FIG. 5 is a device host interface in an embodiment;
FIG. 6 is a real-time serial data interface.
Detailed Description
In order to clarify the technical solution and the working principle of the present invention, the present invention is further described with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1, the power system edge calculation and data distribution device based on the industrial personal computer includes an input/output module, a data acquisition module, a core industrial control module, a data processing module, a state monitoring module, a data distribution module, and a first bus.
The first bus is used as a common channel for transmitting information in the device, and the data processing module, the state monitoring module and the data distribution module are in communication connection with the core industrial control module through the first bus.
The field main body of the electric power system edge calculation and data distribution device is an industrial touch control integrated machine, so that the input/output module is realized by adopting a touch screen, in the embodiment, a 17-inch capacitive touch type LED liquid crystal screen SG17A is preferably selected, and an input unit of the device is connected with a core industrial control module through a man-machine interface, is used for receiving a user operation command and transmits the user operation command to the core industrial control module; and the output unit displays information such as data processing results/state alarm and the like in real time according to the instruction sent by the core industrial control module.
The data acquisition module establishes communication connection with the A, B terminal of the intelligent electric meter through the RS485 interface based on the control instruction of the core industrial control module so as to acquire total power load data sent by the intelligent electric meter. The data acquisition module is provided with a serial port data sending unit, a serial port data receiving unit, a data protocol analysis unit and a data storage unit, and the work flow is as follows:
the serial port data sending unit sends a data packet to the intelligent electric meter according to a communication protocol (DL/T645-2007) between the serial port data sending unit and the intelligent electric meter, data items are represented by using a compressed BCD code, low bytes are in front, high bytes are in back, 33H is added to each byte in each frame data field for encryption, and the format of the data packet is shown in Table 1;
receiving return data from the intelligent ammeter by a serial port data receiving unit;
the data protocol analysis unit analyzes the returned data according to a communication protocol, subtracts 33H from the data field data of the received electric meter returned frame for decryption, and extracts total power load data;
and the data storage unit stores the total power load data for the core industrial control module to fetch.
The core industrial control module comprises a CPU microprocessor, an RAM memory, a memory and a second bus. The second bus is used as a common channel for transmitting information inside the core industrial control module, and the RAM memory, the memory and the I/O interface establish communication connection with the CPU microprocessor through the second bus, in this embodiment:
the type of the RAM memory is 2G/DDR 3;
the memory is a 32G solid state disk and is used for storing total power load data output by the data acquisition module and processing results fed back by the data processing module and the state monitoring module and constructing a system access database;
the CPU microprocessor preferably selects a J1800 dual-core processor, is used for processing total power load data sent by the data acquisition module, respectively transmits corresponding data in the total power load data to the state monitoring module and the data processing module according to a preset program, completes the analysis of the next step, and controls the touch screen to display related information according to processing results fed back by the state monitoring module and the data processing module.
The data processing module comprises a data normalization unit, a power load identification unit and an identification result display unit, and the work flow of the data processing module is as follows:
1) normalizing total power load data (including current, voltage and power) which integrates electricity utilization information of a plurality of types of electric appliances, and setting the load voltage, the load current and the load power as uk、ikAnd pkThen, the normalized load data are respectively:
in the formula, a subscript k represents the data acquisition time, min represents the minimum value of the load quantity, max represents the maximum value of the load quantity, and N is the capacity of the data set;
2) based on the data obtained by continuous monitoring, a load data set data is established, wherein the data set comprises three dimensions of normalized voltage, current and power, namely data ═ u ' i ' p ']Wherein u ', i ', and p ' are the load data u ' at each time 'k、i′k、p′kA composed three-dimensional data set;
3) dividing the load data set into a training set train and a test set test, and establishing a BP neural network algorithm model based on genetic algorithm optimization by using the test set train, as shown in FIG. 3, the process is as follows: according to a coding rule of system design, multi-dimensional data of 3 parameters (voltage, current and power) are randomly generated and used for constructing an initialization population; calculating the fitness of the individual, and continuously generating new individuals by using methods such as selection, intersection, variation and the like, so that the individuals with higher fitness are inherited to the next generation until the algorithm meets the termination condition; after the genetic algorithm part is terminated, decoding the optimal individual into a group of BP neural network connection weight and biased distribution as initial parameters of the BP neural network according to the coding rule; the BP neural network starts from initial parameters, calculates network output errors by using a training data set, reversely transmits the errors and corrects weights and offsets if termination conditions are not met, terminates network learning until the errors meet the conditions, and obtains a final calculation model; FIG. 4 shows the recognition results of the model and the traditional BP neural network model, and it can be known from the figure that the BP neural network algorithm based on genetic algorithm optimization obtains the global optimal solution with a faster convergence rate;
4) adopting a BP neural network algorithm optimized based on a genetic algorithm to identify the power load and determining the type and name of the power load;
5) the identification result display unit is responsible for feeding back the identification result to the core industrial control module and triggering the core industrial control module to control the touch screen to display the identification result in real time.
The state monitoring module comprises a state threshold setting unit, a state monitoring unit and an alarm indicating unit, and the working process of the state monitoring module is as follows:
the state threshold value setting unit completes setting of the threshold value range of the load parameter in the normal state according to the instruction input by the user;
the state monitoring unit compares the received power load data with the load parameter threshold range, and sends a signal to the alarm indicating unit when the load parameter threshold range exceeds the normal state;
and the alarm indication unit feeds back a processing result to the core industrial control module, and controls the touch screen to carry out alarm indication on the load parameter exceeding the normal threshold range.
The data distribution module is used for being connected with the electric power data cloud end and comprises a data sending control unit, a GPRS unit and a remote data receiving unit, and the work flow of the data distribution module is as follows:
the data sending control unit is used for setting sent contents according to an instruction output by a user, extracting data from a system access database, sending the data to the electric power data cloud end through the GPRS unit according to a communication protocol, and can be used for establishing an electric power load big data set;
the remote data receiving unit receives the power load data through the GPRS unit and analyzes the received data according to a communication protocol.
Fig. 5 is a main control interface according to a specific embodiment of the present invention, and the operation method thereof is as follows:
and connecting the device with an RS485 interface of the intelligent electric meter, and electrifying and initializing the system. A user logs in a main control interface of the device, serial port communication setting, acquisition frequency setting, threshold parameter setting and data storage setting are completed in a system functional area of the main control interface, a starting button is clicked, communication between the device and the intelligent electric meter is established, a data acquisition module acquires power load data from the intelligent electric meter based on a communication protocol DL/T645-2007, as shown in fig. 6, the type of the load data depends on use requirements, and the type of total power load data comprises used power, current voltage, current, active power, reactive power and apparent power. The data processing module processes the power load data, and the data processing module comprises data normalization, power load identification and identification result display, wherein the identification result is displayed in an electric appliance monitoring area of the touch screen, the monitoring data can be subjected to statistical analysis, and the monitoring data can be started by clicking a real-time data display button and a load data analysis button in a main control interface system functional area. And the state monitoring module monitors the voltage and the current of the system, and when the voltage/current exceeds the threshold range set by the system, a state alarm button on the touch screen displays red. The user sets the content of the selectable setting transmission data through the data transmission control unit of the data distribution module.
The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to explain the principles of the invention, and that various changes and modifications may be made therein without departing from the spirit and scope of the invention. The scope of the invention is defined by the appended claims, the description and their equivalents.