Method, device, medium and equipment for determining operating mode of generator

Abstract

The present application provides a method, device, medium and equipment for determining the working mode of a generator, which relates to the technical field of generator control, and includes: collecting current data and voltage data of an accessed target load within a first target time window to obtain a target feature vector T; inputting T into a first type recognition model to obtain a first classification result; if the first classification result indicates that the probability of the target load belonging to any load type is less than a preset probability threshold, then obtaining a frequency-domain voltage data matrix PC' and a frequency-domain current data matrix PD'; and obtaining a second classification result according to PC', PD' and a second type recognition model. The working mode finally determined by the present application is more accurate, so as to facilitate timely adjustment of the working mode of the generator, enabling the electric energy to be fully utilized.

Description

Background Art

[0002] In today's power application field, generators, as important power supply devices, are widely used in industrial production, daily life, and various emergency power supply scenarios. When dealing with the matching relationship between generators and loads in the prior art, the main approach is to set different powers according to different loads. This method simply takes power as the only consideration factor and defaults that as long as the load powers are the same, the load-carrying performance of the generator is consistent. However, the actual situation is far more complex. For different loads, even if their powers are the same, due to their corresponding different load types, such as inductive loads, capacitive loads, and resistive loads, etc., there are significant differences in their current characteristics, voltage characteristics, etc. during operation, which in turn lead to greatly different corresponding load-carrying performances. This causes the situation that when the generator faces loads of the same power, some can drive normally while some cannot. This situation not only causes waste of power resources but also greatly limits the effective use range of the generator, and cannot make full and efficient use of electrical energy. There is an urgent need for a new technical solution to solve such problems. Summary of the Invention

[0003] In view of the above technical problems, the present application provides a method, device, medium, and equipment for determining the working mode of a generator, which at least partially solves the problems existing in the prior art.

[0004] In the first aspect of the present application, a method for determining the working mode of a generator is provided. The method includes:

[0005] Collect current data and voltage data of an accessed target load within a first target time window to obtain a target feature vector T=(A, B, C, D), where A is the phase difference; B is the power factor; C is a list of voltage time-domain data; D is a list of current time-domain data; the end time of the first target time window is the current time;

[0006] Input T into a first type recognition model to obtain a first classification result; where the first classification result is used to represent the probability that the target load belongs to each load type; and each load type has a corresponding generator working mode;

[0007] If the probability that the first classification result represents the target load belonging to any load type is less than a preset probability threshold, then convert C' and D' into corresponding frequency-domain voltage data matrix PC' and frequency-domain current data matrix PD' through short-time Fourier transform; where C' and D' are respectively a list of voltage time-domain data and a list of current time-domain data of the accessed target load within a second target time window; the start time of the second target time window is the same as the start time of the first target time window; the time length of the second target time window is greater than the length of the first target time window;

[0008] According to PC’, PD’ and the second type recognition model, a second classification result is obtained; wherein, the second classification result is used to adjust the generator operating mode according to the probability that the target load belongs to each load type.

[0009] In a second aspect of the present application, a device for determining the operating mode of a generator is provided, and the device includes:

[0010] An acquisition unit, configured to acquire current data and voltage data of an accessed target load within a first target time window to obtain a target feature vector T = (A, B, C, D), where A is the phase difference; B is the power factor; C is a list of voltage time-domain data; D is a list of current time-domain data; the end time of the first target time window is the current time;

[0011] A first classification unit, configured to input T into a first type recognition model to obtain a first classification result; wherein, the first classification result is used to represent the probability that the target load belongs to each load type; wherein, each load type has a corresponding generator operating mode;

[0012] A conversion unit, configured to, if the first classification result indicates that the probability that the target load belongs to any load type is less than a preset probability threshold, convert C’ and D’ into a corresponding frequency-domain voltage data matrix PC’ and a frequency-domain current data matrix PD’ through short-time Fourier transform; wherein, C’ and D’ are respectively a list of voltage time-domain data and a list of current time-domain data of the accessed target load within a second target time window; the start time of the second target time window is the same as the start time of the first target time window; the time length of the second target time window is greater than the length of the first target time window;

[0013] A second classification unit, configured to obtain a second classification result according to PC’, PD’ and the second type recognition model; wherein, the second classification result is used to adjust the generator operating mode according to the probability that the target load belongs to each load type.

[0014] In a third aspect of the present application, a non-transitory computer-readable storage medium is provided, and at least one instruction or at least one program segment is stored in the storage medium, and at least one instruction or at least one program segment is loaded and executed by a processor to implement the foregoing method for determining the operating mode of a generator.

[0015] In a fourth aspect of the present application, an electronic device is provided, including a processor and the foregoing non-transitory computer-readable storage medium.

[0016] The present application has at least the following beneficial effects:

[0017] The method for determining the working mode of the generator provided in this application. Since different types of AC loads have different phase differences, power factors, voltage change characteristics, and current change characteristics when first connected to the generator, and these factors affect the load-carrying performance of the AC load, and further affect the working logic of the adapted generator. Judging the load type using only a single characteristic may lead to inaccurate results due to data errors or other factors. Therefore, in this application, the current data and voltage data of the connected target load within the first target time window are collected to obtain the target feature vector T, which includes multi-dimensional data. According to T and the first category recognition model, if the load type to which the target load belongs cannot be determined, it may indicate that the current data volume is small. Furthermore, the second target time window in this application extends for a period of time based on the first time window. Then, when judging the load type again based on the frequency domain data, the original collected data volume is larger, the data dimension is richer, and the time-frequency data after short-time Fourier transform can show how the frequency components of the signal evolve over time on the two-dimensional plane of time and frequency. For example, during the motor startup process, the frequency of the current signal gradually changes from low frequency to the rated frequency, and this frequency change process can be clearly observed through the time-frequency data, while it is relatively difficult to directly distinguish this frequency dynamic change in the time-domain data. In addition, the time-frequency data can more intuitively display the harmonic (frequency components that are integer multiples of the fundamental frequency) and inter-harmonic (frequency components that are non-integer multiples of the fundamental frequency) components in the voltage and current signals and their changes over time. By observing the time-frequency diagram, the occurrence time, duration, and amplitude changes of different-order harmonics and inter-harmonics can be clearly seen. In the power system, harmonics and inter-harmonics have an adverse impact on electrical equipment, and accurate analysis of them is crucial for evaluating the power quality and the operating state of the equipment, while it is difficult for the time-domain data to directly provide these detailed harmonic information. Therefore, in this embodiment, after being converted into time-frequency data, it is easier to analyze the transformation law of the voltage and current signals of the target load current and voltage within a short time during the engine startup compared to the original time-domain data, and the judgment of the load type of the target load can be more accurate and intuitive. This makes the finally determined working mode more accurate, so as to adjust the working mode of the generator in a timely manner and make full use of the electric energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of this application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 It is a flowchart of the method for determining the working mode of the generator provided in the embodiment of this application;

[0020] Figure 2 The block diagram of the working mode determination device for the generator provided by the embodiment of the present application. Detailed implementation manners

[0021] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data used can be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any corresponding deformations are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0023] It should be noted that the following describes various aspects of the embodiments within the scope of the appended claims. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present application, those skilled in the art should understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement a device and / or practice a method. In addition, this device and / or this method can be implemented using other structures and / or functions in addition to one or more of the aspects described herein.

[0024] Please refer to Figure 1 As shown, the embodiment of the present application provides a method for determining the working mode of a generator, and the method includes:

[0025] S100, collect the current data and voltage data of the connected target load within the first target time window to obtain the target feature vector T = (A, B, C, D), where A is the phase difference; B is the power factor; C is the voltage time domain data list; D is the current time domain data list; the end time of the first target time window is the current time.

[0026] Specifically, the generator in this application is a digital generator, which is a relatively advanced small generator. It adopts digital circuit control technology and inverter technology. The working principle of the digital generator is that the engine drives the generator rotor to rotate to generate alternating current first, and this process is similar to that of traditional generators. However, the generated alternating current is not directly output. Instead, it is first converted into direct current by a rectifier, and then an inverter is used to invert the direct current back into high-quality alternating current with stable frequency and voltage, and finally output to the load for use. In this process, the digital circuit control technology precisely adjusts and controls the operation of the inverter to ensure the stability and high quality of the output electric energy. And it has an intelligent control function, which can monitor the operating status of the generator in real time, such as voltage, frequency, speed, fuel quantity, etc., and automatically alarm or stop for protection when abnormalities occur, improving the reliability and safety of the generator. The digital circuit control technology can automatically adjust the speed of the engine according to the size of the actual load, reduce the speed under light load, reduce fuel consumption, and improve fuel economy.

[0027] The digital generator is equipped with a variable-frequency inverter, which can control the working mode of the digital generator to supply different types of AC loads. An AC load refers to various devices or components that consume electric energy or affect the AC power supply in an AC circuit. According to its working principle and characteristics, it can be divided into resistive loads, inductive loads, capacitive loads, and composite loads, etc.

[0028] The working principle of a resistive load is: it works by using the heat generated when current passes through a resistor. Its current and voltage are in the same phase and follow Ohm's law, that is, the magnitude of the current is proportional to the voltage and inversely proportional to the resistance. Common devices such as electric water heaters, electric ovens, incandescent lamps, etc. The heating wires in electric water heaters, the heating tubes in electric ovens, and the filaments in incandescent lamps are all typical resistor elements, which convert electrical energy into heat energy to achieve the purpose of heating or lighting.

[0029] The working principle of an inductive load is: it works based on the principle of electromagnetic induction. When current passes through an inductor coil, a magnetic field will be generated, and the change of the magnetic field will hinder the change of the current, resulting in the current lagging behind the voltage by a certain phase angle. Common devices such as motors, transformers, electromagnets, etc. Taking a motor as an example, the stator winding of the motor is inductive in nature. When operating, it needs to establish a magnetic field to drive the rotor to rotate, and it will consume a certain amount of reactive power, showing the characteristics of an inductive load. Transformers also use the principle of electromagnetic induction to change voltage and transmit electric energy, and their windings also have inductive characteristics.

[0030] The working principle of capacitive load is as follows: It works by utilizing the characteristics of capacitors to store charge and electric field energy. When a capacitor is connected to an AC circuit, the change in voltage causes the capacitor to continuously charge and discharge, thus forming a current. The current leads the voltage by a certain phase angle. Common devices include filter capacitors, compensation capacitors in some electronic circuits, and capacitor motors. In the power supply circuit of electronic devices, filter capacitors are used to smooth the DC voltage and filter out the AC components. Capacitor motors use capacitors to split the phase, causing the stator winding of the motor to generate a rotating magnetic field, thereby driving the rotor to rotate.

[0031] A composite load refers to a load that contains multiple load characteristics. The relationship between its current and voltage is relatively complex. It may have both a resistive energy-consuming part and an inductive or capacitive energy-storing part. Common devices include televisions, computers, etc. These devices have both resistive components for heating inside, such as current-limiting resistors in the power supply part, and a large number of capacitor and inductor components for filtering, signal processing, etc. Their overall load characteristics are a combination of multiple characteristics.

[0032] In terms of the phase of current and voltage, since the phase of the current and voltage of a resistive load is the same. That is, when the voltage reaches its peak value, the current also reaches its peak value; when the voltage is zero, the current is also zero. By measuring the current and voltage signals through a sensor, if the phase difference between the two is analyzed to be 0°, it can be basically judged as a resistive load. For an inductive load, the current lags behind the voltage. In an AC circuit, the inductor will impede the change in current, causing the change in current to always lag behind the change in voltage. If the sensor detects that the current phase lags behind the voltage phase, such as lagging by 30°, 60°, or even 90° (theoretically lagging by 90° in the case of a pure inductor), it can be judged as an inductive load. For a capacitive load, the current leads the voltage. Because during the charging and discharging process of the capacitor, the change in current will precede the change in voltage. When the sensor detects that the current phase leads the voltage phase, such as leading by 30°, 60°, or approaching 90° (theoretically leading by 90° in the case of a pure capacitor), it can be judged as a capacitive load.

[0033] In terms of the waveform characteristics of current and voltage, the current and voltage waveforms of resistive loads are usually standard sine waves, consistent with the AC waveforms output by the inverter, and the waveforms are smooth without obvious distortion or harmonic components. Due to the presence of inductance, during the startup and operation of inductive loads, the current waveform may exhibit certain delays and deformations. For example, when a motor starts, the current will increase instantaneously, and the waveform may have spikes or be uneven. During operation, due to factors such as the magnetic field changes inside the motor, the current waveform may contain certain harmonic components, making the waveform no longer a standard sine wave. For capacitive loads, the current waveform also changes during the charging and discharging processes. When the capacitor is charging, the current will rise rapidly and then gradually decrease; when discharging, the direction of the current is opposite. Therefore, the current waveform of capacitive loads may be asymmetric or have obvious sudden changes, which is significantly different from the waveforms of resistive and inductive loads.

[0034] Furthermore, the power factor is an important parameter for measuring the ability of loads in a power system to effectively convert electrical energy into useful work. In this application, the power factor is calculated through the following steps:

[0035] S110, calculate the effective voltage U.

[0036] Here, C = (C1, C2,..., C i ,..., C n ); i = 1, 2,..., n; n is the number of sampling time points within the first target time window; the time interval between any two samplings is the same; C i is the voltage value collected at the i-th sampling time point within the first target time window; D = (D1, D2,..., D i ,..., D n ); D i is the current value collected at the i-th sampling time point within the first target time window. If the voltage is a standard sine wave, assuming the maximum voltage within the first time window is U m , according to the relationship between the effective value and the maximum value of sinusoidal AC voltage, the formula can be used; if it is a non-standard sine wave, then U conforms to the following characteristics:

[0037]

[0038] S120, calculate the effective current I.

[0039] Here, for sinusoidal AC current, if the maximum current collected is I m , then the effective current is I = I m / √2; if it is a non-standard sine wave, then .

[0040] S130. Obtain the apparent power S based on U and I, where S = UI, and the unit is volt-ampere (VA).

[0041] S140. Calculate the active power P.

[0042] Here, .

[0043] S150. Calculate the power factor cosψ, where cosψ = P / S. The power factor is a dimensionless value, and its value range is between 0 and 1.

[0044] The power factor of resistive loads is usually close to 1 because resistive loads only consume active power and hardly consume reactive power. The power factor of inductive loads is generally less than 1 and is usually between 0.5 - 0.8, depending on the size of the inductor and the characteristics of the load. This is because inductive loads need to consume a certain amount of reactive power to establish a magnetic field. The power factor of capacitive loads is also less than 1, but different from inductive loads, the reactive power of capacitive loads is output outward, and the angle of its power factor is leading, while that of inductive loads is lagging. Generally, the power factor of capacitive loads is between 0.6 - 0.9.

[0045] In summary, due to the different phase differences, power factors, voltage change characteristics, and current change characteristics of different types of AC loads, using only a single characteristic to judge the load type may lead to inaccurate results due to data errors or other factors. Therefore, in this application, the current data and voltage data of the connected target load within the first target time window are collected to obtain the target feature vector T = (A, B, C, D). Here, both C and D are discrete sequence data, which can better reflect the voltage and current changes of the connected target load within the first target time window. Additionally, in this application, n = L×E, where L is the time length of the first time window and E is the sampling rate. Further, L = 2 seconds and E ≥ 1 MHz. That is, the current and voltage data obtained in this application are those within a short time after the target load is connected, and the corresponding target feature vector within the first target time window is obtained therefrom. This is because within a short time after the AC load is connected to the power supply, the current or voltage may change significantly, and the obtained data is more accurate for judging the change of the load type. Additionally, in order to achieve precise control of the digital generator frequency conversion inverter, it is necessary to judge the AC load type within a short time to switch to the working mode of the generator corresponding to the current AC load type, so that the target load can work efficiently, avoid power loss, and improve the power utilization efficiency.

[0046] S200, input T into the first type of recognition model to obtain a first classification result; wherein, the first classification result is used to represent the probability that the target load belongs to each load type; wherein, each load type has a corresponding generator operating mode.

[0047] Specifically, the first type of recognition model is pre-trained to classify the input feature vector and obtain the probability that the input feature vector belongs to each load type. It should be noted that the first type of recognition model can be any model well-known to those skilled in the art that can achieve the purpose of multi-classification.

[0048] S300, if the probability that the first classification result represents the target load belonging to any load type is less than the preset probability threshold, then convert C' and D' into corresponding frequency-domain voltage data matrix PC' and frequency-domain current data matrix PD' through short-time Fourier transform; wherein, C' and D' are respectively the voltage time-domain data list and current time-domain data list of the connected target load within the second target time window; the start time of the second target time window is the same as the start time of the first target time window; the time length of the second target time window is greater than the length of the first target time window.

[0049] Specifically, the working process of the first type of recognition model is as follows: if the probability that the input feature vector belongs to any load type is greater than the preset probability threshold, then take this load type as the output result. On the contrary, if the probability that the input feature vector belongs to each load type is less than the preset probability threshold, it means that the input feature vector may not belong to any load type. In this application, it may be because the sampling time is too short, resulting in the first type of recognition model being unable to accurately judge the load type based on the current data. Therefore, in this embodiment, when the probability that the first classification result represents the target load belonging to any load type is less than the preset probability threshold, that is, the first type of recognition model is unable to accurately judge the load type based on the current data, convert C' and D' into corresponding frequency-domain voltage data matrix PC' and frequency-domain current data matrix PD' through short-time Fourier transform. The short-time Fourier transform can divide the data into multiple short time periods and perform Fourier transform on each short time period respectively, which can display the time and frequency information of the signal at the same time and form time-frequency data. In this way, the change of different frequency components in the voltage data and current data over time within the second target time window can be clearly observed.

[0050] In an exemplary embodiment of the present application, PC' conforms to the following characteristics:

[0051]

[0052] x = 1, 2,..., y; y is the number of frequency points; p = 1, 2,..., q; q is the number of points on the time axis; PC'1，p For the amplitude and phase information of the voltage signal at the first frequency point corresponding to the p-th time point on the time axis; PC’ x，p For the amplitude and phase information of the voltage signal at the x-th frequency point corresponding to the p-th time point on the time axis.

[0053] Furthermore, similar to the above, PD’ conforms to the following characteristics:

[0054]

[0055] PD’ 1，p For the amplitude and phase information of the current signal at the first frequency point corresponding to the p-th time point on the time axis; PD’ x，p For the amplitude and phase information of the current signal at the x-th frequency point corresponding to the p-th time point on the time axis.

[0056] It should be noted that the second target time window is extended by a period of time based on the first time window because the first type recognition model is not sufficient to accurately judge the load type according to the current data. Therefore, in this embodiment, when the load type is judged again based on the frequency domain data, the original collected data volume is larger, the data dimension is richer, and the time-frequency data after short-time Fourier transform can show how the frequency components of the signal evolve over time on the two-dimensional plane of time and frequency. For example, during the motor startup process, the frequency of the current signal will gradually change from low frequency to the rated frequency, and this frequency change process can be clearly observed through the time-frequency data, while it is relatively difficult to directly distinguish this frequency dynamic change in the time-domain data. In addition, the time-frequency data can more intuitively display the harmonic (frequency components that are integer multiples of the fundamental frequency) and inter-harmonic (frequency components that are non-integer multiples of the fundamental frequency) components in the voltage and current signals and their changes over time. By observing the time-frequency diagram, the appearance time, duration, and amplitude changes of different harmonics and inter-harmonics can be clearly seen. In the power system, harmonics and inter-harmonics will have an adverse impact on electrical equipment, and accurate analysis of them is crucial for evaluating the power quality and equipment operation status, while it is difficult for time-domain data to directly provide these detailed harmonic information. Therefore, in this embodiment, after being converted into time-frequency data, it is easier to analyze the transformation law of the voltage and current signals of the target load current and voltage within a short time during the engine startup compared to the original time-domain data, and the judgment of the load type of the target load can be more accurate and intuitive.

[0057] S400. Obtain a second classification result according to PC’, PD’ and the second type recognition model; wherein, the second classification result is used to adjust the generator working mode according to the probability of the target load belonging to each load type.

[0058] Specifically, step S400 includes:

[0059] S410. Obtain the corresponding voltage frequency-domain diagram APC' and current frequency-domain diagram APD' according to PC' and PD'.

[0060] S420. Respectively perform image feature extraction on APC' and APD' according to the CNN model to obtain a voltage frequency-domain feature vector APCT' and a current frequency-domain feature vector APDT'.

[0061] S430. Perform feature fusion on APCT' and APDT' to obtain a fused frequency-domain feature vector R.

[0062] S440. Input R into the second type of recognition model to obtain a second classification result.

[0063] In this embodiment, all images are adjusted to the size required by the CNN model, and then the pixel values of the images are normalized, and the pixel values are scaled to a fixed range, which helps the convergence of the model. And in order to increase the diversity of the data set and the generalization ability of the model, some data augmentation operations can be performed, such as random cropping, flipping, rotation, etc. The feature vectors of the voltage image and the current image are concatenated in the feature dimension to form a longer feature vector. Finally, a second classification result is obtained.

[0064] Here, the second classification result is used to output the load type corresponding to the target load with the highest probability and greater than the preset probability threshold according to the probability of the target load belonging to each load type. In this embodiment, the variable-frequency inverter is controlled to adjust the working mode of the generator to the working mode corresponding to the load type to which the target load belongs. So that the target load works efficiently, avoiding power loss, and improving the power utilization efficiency. Here, the working mode of the generator is the control logic of the generator. This application only needs to adjust the control logic of the generator to make full and efficient use of electric energy without adjusting the circuit.

[0065] In an exemplary embodiment of the present application, the method further includes:

[0066] If the probability that the first classification result represents that the target load belongs to any load type is equal to or greater than the preset probability threshold, then use this load type as the target load type corresponding to the target load.

[0067] In an exemplary embodiment of the present application, after respectively performing image feature extraction on APC' and APD' according to the CNN model to obtain a voltage frequency-domain feature vector APCT' and a current frequency-domain feature vector APDT', the method further includes:

[0068] S450. Perform feature fusion on C', D', APCT' and APDT' to obtain a fused frequency-domain feature vector R'.

[0069] S460, input R' into the second type of recognition model to obtain the second classification result.

[0070] In this embodiment, the original voltage time-domain data and current time-domain data, as well as the voltage frequency-domain data and current frequency-domain data obtained after short-time Fourier transform, are jointly spliced to obtain the fused frequency-domain feature vector R'. Here, adding the original time-domain data enables the final model to consider both time-domain features and frequency-domain features, so that the obtained classification result is more accurate.

[0071] Please refer to Figure 2 As shown, an embodiment of the present application provides a device 100 for determining the working mode of a generator. The device includes:

[0072] An acquisition unit 110, configured to acquire current data and voltage data of an accessed target load within a first target time window to obtain a target feature vector T = (A, B, C, D), where A is the phase difference; B is the power factor; C is a list of voltage time-domain data; D is a list of current time-domain data; the end time of the first target time window is the current time.

[0073] A first classification unit 120, configured to input T into a first type of recognition model to obtain a first classification result; where the first classification result is used to represent the probability that the target load belongs to each load type; and each load type has a corresponding generator working mode.

[0074] A conversion unit 130, configured to, if the first classification result indicates that the probability that the target load belongs to any load type is less than a preset probability threshold, convert C' and D' into corresponding frequency-domain voltage data matrix PC' and frequency-domain current data matrix PD' through short-time Fourier transform; where C' and D' are respectively a list of voltage time-domain data and a list of current time-domain data of the accessed target load within a second target time window; the start time of the second target time window is the same as the start time of the first target time window; the time length of the second target time window is greater than the length of the first target time window.

[0075] A second classification unit 140, configured to obtain a second classification result according to PC', PD' and a second type of recognition model; where the second classification result is used to adjust the generator working mode according to the probability that the target load belongs to each load type.

[0076] In an exemplary embodiment of the present application, an electronic device capable of implementing the above method is further provided.

[0077] Those skilled in the art can understand that various aspects of the present application can be implemented as a system, a method, or a program product. Therefore, various aspects of the present application can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which can be collectively referred to as "circuits", "modules", or "systems" here.

[0078] An electronic device according to this embodiment of the present application. The electronic device is only an example and should not impose any limitation on the functions and usage scope of the embodiments of the present application.

[0079] The electronic device is presented in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: the above-mentioned at least one processor, the above-mentioned at least one storage, and a bus connecting different system components (including the storage and the processor).

[0080] Among them, the storage stores program code, and the program code can be executed by the processor, so that the processor executes the steps according to various exemplary embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0081] The storage may include a readable medium in the form of a volatile storage, such as a random access storage (RAM) and / or a cache storage, and may further include a read-only storage (ROM).

[0082] The storage may further include a program / utility with a set (at least one) of program modules. Such program modules include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include the implementation of a network environment.

[0083] The bus may represent one or more of several types of bus structures, including a memory bus or a memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any bus structure in a variety of bus structures.

[0084] The electronic device can also communicate with one or more external devices (such as a keyboard, a pointing device, a Bluetooth device, etc.), and can also communicate with one or more devices that enable a user to interact with the electronic device, and / or communicate with any device that enables the electronic device to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication can be carried out through an input / output (I / O) interface. Moreover, the electronic device can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter. As shown in the figure, the network adapter communicates with other modules of the electronic device through a bus. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in combination with the electronic device, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0085] Through the description of the above embodiments, those skilled in the art can easily understand that the exemplary embodiments described herein can be implemented by software, or can be implemented by the way of software combined with necessary hardware. Therefore, the technical solutions according to the embodiments of the present application can be embodied in the form of a software product, and the software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present application.

[0086] In an exemplary embodiment of the present application, there is also provided a computer-readable storage medium, on which a program product capable of implementing the above method of this specification is stored. In some possible implementation manners, various aspects of the present application can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0087] The program product can adopt any combination of one or more readable media. The readable media can be a readable signal medium or a readable storage medium. The readable storage medium can, for example, but not be limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0088] A computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which readable program code is carried. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The readable signal medium may also be any readable medium other than a readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.

[0089] The program code contained on the readable medium may be transmitted by any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0090] The program code for performing the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's device, executed as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., by using an Internet service provider to connect through the Internet).

[0091] In addition, the above drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present application, and are not for limiting purposes. It is easy to understand that the processes shown in the above drawings do not indicate or limit the chronological order of these processes. Additionally, it is also easy to understand that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0092] It should be noted that although several modules or units of devices for action execution are mentioned in the above detailed description, such a division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more of the above-described modules or units may be embodied in one module or unit. Conversely, the features and functions of one module or unit described above may be further divided and embodied by multiple modules or units.

[0093] The above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present application should be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for determining a working mode of a generator, characterized in that: The method comprises: Collect the current data and voltage data of the connected target load in the first target time window to obtain the target feature vector T=(A, B, C, D), where A is the phase difference; B is the power factor; C is the voltage time domain data list; D is the current time domain data list; the end time of the first target time window is the current time; Input T into the first type recognition model to obtain a first classification result; wherein the first classification result is used to characterize the probability that the target load belongs to each load type; wherein each load type has a corresponding generator working mode; the generator is a digital generator; the digital generator is provided with a variable frequency inverter, and the variable frequency inverter is used to control the working mode of the digital generator; If the probability that the first classification result indicates that the target load belongs to any load type is less than the preset probability threshold, C' and D' are converted into the corresponding frequency domain voltage data matrix PC' and frequency domain current data matrix PD' through short-time Fourier transform; wherein C' and D' are respectively the voltage time domain data list and the current time domain data list of the connected target load in the second target time window; the start time of the second target time window is the same as the start time of the first target time window; the time length of the second target time window is greater than the length of the first target time window; According to PC', PD', and the second type identification model, a second classification result is obtained; wherein the second classification result is used to determine the generator operating mode according to the probability that the target load belongs to each load type; C=(C1,C2,…,C i , …, C n ); i = 1, 2, ..., n; n is the number of sampling time points in the first target time window; the time interval between any two samplings is the same; C i is the voltage value collected at the i-th sampling time point in the first target time window; D=(D1, D2, …, D i , …, D n );D i The current value collected at the i-th sampling time point in the first target time window; n=L×E; where L is the time length of the first time window; E is the sampling rate; the time length L of the first time window is 2 seconds; the sampling rate E≥1MHz; According to PC', PD' and the second type recognition model, the second classification result is obtained, including: According to PC' and PD', the corresponding voltage frequency domain diagram APC' and current frequency domain diagram APD' are obtained; According to the CNN model, image features of APC' and APD' are extracted to obtain voltage frequency domain feature vector APCT' and current frequency domain feature vector APDT'; Perform feature fusion on APCT' and APDT' to obtain the fused frequency domain feature vector R; R is input into the second type recognition model to obtain a second classification result.

2. The method for determining the working mode of a generator according to claim 1, characterized in that: The method further comprises: If the first classification result indicates that the probability that the target load belongs to any load type is equal to or greater than a preset probability threshold, then the load type is used as the target load type corresponding to the target load.

3. The method for determining the working mode of a generator according to claim 1, characterized in that: After performing image feature extraction on APC' and APD' according to the CNN model to obtain a voltage frequency domain feature vector APCT' and a current frequency domain feature vector APDT', the method further includes: Perform feature fusion on C', D', APCT' and APDT' to obtain the fused frequency domain feature vector R'; R' is input into the second type recognition model to obtain a second classification result.

4. A device for determining a working mode of a generator, characterized in that: The device comprises: A collection unit is used to collect current data and voltage data of the connected target load within the first target time window to obtain a target feature vector T=(A, B, C, D), wherein A is the phase difference; B is the power factor; C is the voltage time domain data list; D is the current time domain data list; and the end time of the first target time window is the current time; A first classification unit is used to input T into a first type recognition model to obtain a first classification result; wherein the first classification result is used to characterize the probability that the target load belongs to each load type; wherein each load type has a corresponding generator working mode; the generator is a digital generator; the digital generator is provided with a variable frequency inverter, and the variable frequency inverter is used to control the working mode of the digital generator; A conversion unit, for converting C' and D' into corresponding frequency domain voltage data matrix PC' and frequency domain current data matrix PD' by short-time Fourier transform if the probability represented by the first classification result that the target load belongs to any load type is less than a preset probability threshold; wherein C' and D' are respectively a voltage time domain data list and a current time domain data list of the connected target load within a second target time window; the start time of the second target time window is the same as the start time of the first target time window; and the time length of the second target time window is greater than the length of the first target time window; The second classification unit is used to obtain a second classification result according to PC', PD' and the second type identification model; wherein the second classification result is used to adjust the generator working mode according to the probability that the target load belongs to each load type; C=(C1, C2, ..., C i , …, C n ); i = 1, 2, ..., n; n is the number of sampling time points in the first target time window; the time interval between any two samplings is the same; C i is the voltage value collected at the i-th sampling time point in the first target time window; D=(D1, D2, …, D i , …, D n );D i The current value collected at the i-th sampling time point in the first target time window; n=L×E; where L is the time length of the first time window; E is the sampling rate; the time length L of the first time window is 2 seconds; the sampling rate E≥1MHz; The second classification unit obtains the second classification result according to the following steps: According to PC' and PD', the corresponding voltage frequency domain diagram APC' and current frequency domain diagram APD' are obtained; According to the CNN model, image features of APC' and APD' are extracted to obtain voltage frequency domain feature vector APCT' and current frequency domain feature vector APDT'; Perform feature fusion on APCT' and APDT' to obtain the fused frequency domain feature vector R; R is input into the second type recognition model to obtain a second classification result.

5. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the method for determining the working mode of the generator as described in any one of claims 1-3.

6. An electronic device, characterized in that: The invention comprises a processor and the non-transitory computer-readable storage medium as claimed in claim 5.

Images