Inverter voice interaction system

Abstract

The invention discloses an inverter voice interaction system, which comprises a voice acquisition and conversion module for acquiring a voice signal of a user and converting the voice signal into an electric signal; the voice processing module is used for performing voice recognition and semantic analysis on the electric signal, outputting a voice analog signal according to a semantic analysis result, generating a communication instruction as required, communicating with the inversion control module and outputting the voice analog signal according to communication data; the method has the beneficial effects that core functions of directly controlling the inverter to be turned on and turned off, switching the working frequency, turning on and off at regular time, modifying parameters and the like through voice instructions are supported, the constraint of traditional key operation is eliminated, the method is particularly suitable for quick operation in an off-network scene, and the operation efficiency is remarkably improved; professional settings such as over-voltage and under-voltage values, voltage and current calibration values and overload protection parameters can be directly modified through voice, and the parameter adjustment process is simplified; the working state of the inverter is broadcasted in real time, a user can obtain key data without checking a display screen, and the information obtaining efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of inverter control, and in particular relates to an inverter voice interaction system. Background Art

[0002] An inverter is a power electronic device that converts direct current into alternating current. It plays a key role in many fields. Its core function is to realize the conversion of electrical energy forms to meet the needs of different electrical equipment. According to the output waveform, it can be divided into square wave inverters, modified sine wave inverters and sine wave inverters. According to the application scenario, it includes off-grid inverters and grid-connected inverters. Off-grid inverters are mainly used in independent power supply systems, such as off-grid solar power generation systems, RV power systems, etc., to provide stable AC power for load equipment; grid-connected inverters convert direct current generated by renewable energy sources such as solar energy and wind energy into AC power with the same frequency and phase as the power grid, and feed it into the power grid to realize grid-connected sales of electricity or self-generation and self-use, and surplus power to the grid.

[0003] Patent publication number CN109389977A discloses a voice interaction method and device, which includes responding to a wake-up word input by a user, waking up the voice interaction function of the smart device according to the wake-up word; receiving voice data to be recognized; performing semantic analysis on the voice data to obtain a first semantic instruction, and performing semantic analysis on the voice data in combination with the wake-up word to obtain a second semantic instruction; judging whether the semantic quality of the first semantic instruction is higher than the semantic quality of the second semantic instruction; if the semantic quality of the first semantic instruction is higher than the semantic quality of the second semantic instruction, executing the first semantic instruction; if the semantic quality of the second semantic instruction is higher than the semantic quality of the first semantic instruction, executing the second semantic instruction. However, this technology mainly focuses on voice interaction of general smart devices, and is insufficiently designed for professional function interaction of inverters, and cannot fully meet the diverse voice control needs of inverters in different application scenarios. Summary of the Invention

[0004] The purpose of the present invention is to provide an inverter voice interaction system to achieve more comprehensive and professional voice control of the inverter, thereby improving the convenience and experience of users in using the inverter.

[0005] To achieve the above object, the present invention provides the following technical solution: an inverter voice interaction system, comprising

[0006] Voice collection and conversion module, which collects the user's voice signal and converts it into an electrical signal;

[0007] The voice processing module performs voice recognition and semantic analysis on the electrical signal, outputs a voice analog signal based on the semantic analysis results, generates communication instructions as needed, communicates with the inverter control module, and outputs a voice analog signal based on the communication data;

[0008] The inverter control module collects and processes inverter data, controls the inverter to perform AC / DC conversion, communicates with the voice processing module, and sends communication instructions;

[0009] The voice broadcast module broadcasts according to the signal of the voice processing module;

[0010] Storage module, storing the setting parameters of the inverter;

[0011] Power module, which provides power to the system.

[0012] As a preferred technical solution of the present invention, the voice processing module adopts a neural network intelligent voice chip.

[0013] As a preferred technical solution of the present invention, the working data includes whether the inverter is working, input voltage, output voltage, output power, internal temperature, and fan working status.

[0014] As a preferred technical solution of the present invention, it also includes a display module, which displays the working status and parameters of the inverter.

[0015] As a preferred technical solution of the present invention, the setting parameters of the inverter are stored, including overvoltage and undervoltage values, voltage, current, power and electricity calibration values, overload times, overload value, delay time, and historical working data.

[0016] As a preferred technical solution of the present invention, the user's voice signal is collected and converted into an electrical signal. The specific implementation method is as follows:

[0017] Use high-sensitivity electret microphone or MEMS microphone;

[0018] The weak analog electrical signal output by the microphone is amplified by an operational amplifier, and the high-frequency noise is removed by an RC filter circuit, thereby increasing the signal-to-noise ratio to over 60dB.

[0019] The amplified and filtered analog electrical signal is connected to the ADC, and the analog signal is converted into a digital electrical signal through the sampling theorem;

[0020] The converted digital electrical signal is temporarily stored in the FIFO buffer built into the FLASH or MCU, waiting to be transmitted to the voice processing module.

[0021] As a preferred technical solution of the present invention, the audio signal is converted into a digital electrical signal through a high-performance, low-power audio ADC with an SNR of ≥ 95dB.

[0022] As a preferred technical solution of the present invention, a communication instruction is generated according to the semantic analysis result, and the specific implementation method is as follows:

[0023] According to the inverter control protocol, the structured semantic instructions are mapped into the corresponding communication protocol format;

[0024] Perform type conversion and encoding on parameters in semantic instructions;

[0025] Calculate CRC check code or parity bit according to protocol requirements;

[0026] The mapped protocol data, parameter code and check code are encapsulated into a complete communication frame according to the protocol format and sent to the inverter through the communication module.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] Supports voice commands to directly control the inverter's core functions such as power on and off, switching operating frequency, and timing operation, breaking away from the constraints of traditional button operations. It is especially suitable for quick operation in off-grid scenarios, significantly improving operational efficiency. Professional settings such as overvoltage and undervoltage values, voltage and current calibration values, and overload protection parameters can be directly modified through voice, simplifying the parameter adjustment process.

[0029] Real-time broadcast of the inverter's working status allows users to obtain key data without having to check the display, improving information acquisition efficiency;

[0030] Adopting neural network voice chip, through MFCC feature extraction and LSTM model, the voice recognition accuracy is high in noisy environments, reducing the triggering of false commands;

[0031] It can not only proactively report faults, interact with real-time information, the working status of the inverter, turn the machine on and off, but also modify the internal parameters of the inverter and repair faults autonomously. It is very flexible and clever. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is the interactive principle diagram of the present invention;

[0033] Figure 2 This is a block diagram of the inverter principle of the present invention;

[0034] Figure 3 is a system block diagram of the present invention;

[0035] Figure 4 It is the installation position of each component of the front panel of the inverter of the present invention; DETAILED DESCRIPTION

[0036] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0037] Example 1

[0038] See also Figure 1-Figure 3 , which is the first embodiment of the present invention, provides an inverter voice interaction system, including

[0039] Voice collection and conversion module, which collects the user's voice signal and converts it into an electrical signal;

[0040] The voice processing module performs voice recognition and semantic analysis on the electrical signal, outputs a voice analog signal based on the semantic analysis results, generates communication instructions as needed, communicates with the inverter control module, and outputs a voice analog signal based on the communication data;

[0041] The inverter control module collects and processes inverter data, including whether the inverter is working, input voltage, output voltage, output power, internal temperature, and fan working status, controls the inverter to perform AC / DC conversion, communicates with the voice processing module, and sends communication instructions;

[0042] The voice broadcast module broadcasts according to the signal of the voice processing module;

[0043] Storage module, which stores the inverter setting parameters, including fan start and stop values, voltage, current, power and electricity calibration values, overload times, overload value, delay time, and historical working data;

[0044] Power module, which provides power to the system.

[0045] In this embodiment, preferably, the user's voice signal is collected and converted into an electrical signal. The specific implementation method is as follows:

[0046] Use high-sensitivity electret microphone or MEMS microphone;

[0047] The weak analog electrical signal output by the microphone is amplified by an operational amplifier, and the high-frequency noise is removed by an RC filter circuit, thereby increasing the signal-to-noise ratio to over 60dB.

[0048] The amplified and filtered analog electrical signal is connected to the ADC, and the analog signal is converted into a digital electrical signal through the sampling theorem;

[0049] The converted digital electrical signal is temporarily stored in the FIFO buffer built into the FLASH or MCU, waiting to be transmitted to the voice processing module.

[0050] In this embodiment, preferably, the data is converted into a digital electrical signal through a high-performance, low-power audio ADC with an SNR of ≥ 95dB.

[0051] In this embodiment, preferably, a communication instruction is generated according to the semantic analysis result, and the specific implementation method is as follows:

[0052] According to the inverter control protocol, the structured semantic instructions are mapped into the corresponding communication protocol format;

[0053] Perform type conversion and encoding on parameters in semantic instructions;

[0054] Calculate CRC check code or parity bit according to protocol requirements;

[0055] The mapped protocol data, parameter code and check code are encapsulated into a complete communication frame according to the protocol format and sent to the inverter through the communication module.

[0056] The usage process is as follows:

[0057] The voice acquisition and conversion module uses a high-sensitivity microphone to clearly capture the user's voice signal; the voice processing module uses an advanced neural network intelligent voice chip with powerful voice recognition and semantic analysis capabilities; the inverter control module uses a stable communication protocol to ensure the accuracy and real-time nature of command transmission;

[0058] When the user needs to turn on the inverter, he / she issues a wake-up command of "Hello Xiaoni". The microphone collects the voice signal and converts it into an electrical signal. The voice processing module recognizes the wake-up command and performs semantic analysis to determine it is a wake-up command. The voice processing module waits for the reception and recognition of the command instruction. The user issues a voice command of "turn on the power". The voice processing module determines it as a power-on command and generates a power-on communication instruction, which is transmitted to the inverter through communication data. The inverter performs the power-on operation. At the same time, the inverter control module collects the working data of the inverter, such as input voltage, output voltage, power, working status, fan start and stop, etc., generates a communication instruction, and transmits it to the voice processing module. If necessary, the user can issue a "live broadcast" command. The voice processing module determines it as a live broadcast instruction and generates a voice broadcast simulation signal to broadcast the power-on and power-off status and working data of the inverter.

[0059] When the user needs to modify the overvoltage or undervoltage value, a voice command of "overvoltage modification" is issued, and the voice processing module recognizes and replies "Please say, for example, add one volt, or subtract 0.5 volts"; the user gives an instruction according to the example entry, such as issuing "add one volt", and the voice processing module recognizes and replies "Please confirm, reply, confirm, or cancel", and the user gives an instruction according to the example entry, such as issuing "OK", and the voice processing module recognizes and replies "Confirming" and generates a communication instruction and transmits it to the inverter control module. The inverter control module receives the communication, stores the overvoltage value in the storage module, and responds to the voice processing module with the currently modified overvoltage value. The voice processing module receives the communication data and sends an analog signal to the voice broadcast module, such as broadcasting "The current battery overvoltage value is 16.5 volts";

[0060] In terms of timing work, the user issues a voice command of "scheduled power on", and the voice processing module recognizes and replies "OK, first tell me the number of hours, such as three hours; then tell me the number of minutes, such as ten minutes"; the user gives a command according to the example words, such as "30 minutes", and the voice processing module recognizes and replies "OK", and generates a communication command and transmits it to the inverter control module. After receiving the communication, the inverter control module waits 20 seconds to confirm whether there is another time. If not, it sends a communication command to the voice processing module to confirm the time. After receiving the communication data, the voice processing module sends an analog signal to the voice broadcast module, such as broadcasting "The power on delay of 30 minutes is set."

[0061] Working principle of the inverter circuit: After the inverter main control circuit detects that the battery is connected and recognizes that the battery voltage is normal, it sends a push-pull drive signal to increase the battery voltage to the bus voltage. After detecting that the bus voltage is normal, the inverter main control circuit sends a PWM inverter drive signal to convert the DC carrier into AC power for use by the AC load.

[0062] Working principle of the voice circuit: The microphone converts the language into electrical signals and enters the voice circuit. The voice chip samples through the analog sampling port. The chip has a built-in voice database and recognizes the input voice through comparison. After recognizing the language, if there is a set response voice, it is output to the speaker for broadcast through the chip's digital-to-analog conversion port. The corresponding data can also be sent to other chips through the chip's built-in serial port communication.

[0063] Example 2

[0064] See also Figures 1-4 , which is a second embodiment of the present invention, provides an inverter voice interaction system, including

[0065] Voice collection and conversion module, which collects the user's voice signal and converts it into an electrical signal;

[0066] The voice processing module performs voice recognition and semantic analysis on the electrical signal, outputs a voice analog signal based on the semantic analysis results, generates communication instructions as needed, communicates with the inverter control module, and outputs a voice analog signal based on the communication data;

[0067] The inverter control module collects and processes inverter data, including whether the inverter is working, input voltage, output voltage, output power, internal temperature, and fan working status, controls the inverter to perform AC / DC conversion, communicates with the voice processing module, and sends communication instructions;

[0068] The voice broadcast module broadcasts according to the signal of the voice processing module;

[0069] Storage module, which stores the inverter setting parameters, including over-voltage and under-voltage values, voltage, current, power and electricity calibration values, overload times, overload values, delay time, and historical working data;

[0070] Power module, which provides power to the system.

[0071] In this embodiment, preferably, a display module is further included to display information such as the working status and parameter settings of the inverter, so that the user can intuitively understand the status of the inverter.

[0072] In this embodiment, preferably, the user's voice signal is collected and converted into an electrical signal. The specific implementation method is as follows:

[0073] Use high-sensitivity electret microphone or MEMS microphone;

[0074] The weak analog electrical signal output by the microphone is amplified by an operational amplifier, and the high-frequency noise is removed by an RC filter circuit, thereby increasing the signal-to-noise ratio to over 60dB.

[0075] The amplified and filtered analog electrical signal is connected to the ADC, and the analog signal is converted into a digital electrical signal through the sampling theorem;

[0076] The converted digital electrical signal is temporarily stored in the FIFO buffer built into the FLASH or MCU, waiting to be transmitted to the voice processing module.

[0077] Converted into digital electrical signals through high-performance, low-power audio ADC with SNR ≥ 95dB.

[0078] In this embodiment, preferably, a communication instruction is generated according to the semantic analysis result, and the specific implementation method is as follows:

[0079] According to the inverter control protocol, the structured semantic instructions are mapped into the corresponding communication protocol format;

[0080] Perform type conversion and encoding on parameters in semantic instructions;

[0081] Calculate CRC check code or parity bit according to protocol requirements;

[0082] The mapped protocol data, parameter code and check code are encapsulated into a complete communication frame according to the protocol format and sent to the inverter through the communication module.

[0083] The usage process is as follows:

[0084] The voice acquisition and conversion module uses a high-sensitivity microphone to clearly capture the user's voice signal; the voice processing module uses an advanced neural network intelligent voice chip with powerful voice recognition and semantic analysis capabilities; the communication module uses a stable communication protocol to ensure the accuracy and real-time nature of command transmission;

[0085] When the user needs to turn on the inverter, he / she issues a wake-up command of "Hello Xiaoni". The microphone collects the voice signal and converts it into an electrical signal. The voice processing module recognizes the wake-up command and performs semantic analysis to determine it is a wake-up command. The voice processing module waits for the reception and recognition of the command instruction. The user issues a voice command of "turn on the power". The voice processing module determines it as a power-on command and generates a power-on communication instruction, which is transmitted to the inverter through communication data. The inverter performs the power-on operation. At the same time, the inverter control module collects the working data of the inverter, such as input voltage, output voltage, power, working status, fan start and stop, etc., generates a communication instruction, and transmits it to the voice processing module. If necessary, the user can issue a "live broadcast" command. The voice processing module determines it as a live broadcast instruction and generates a voice broadcast simulation signal to broadcast the power-on and power-off status and working data of the inverter.

[0086] When the user needs to modify the overvoltage or undervoltage value, a voice command of "overvoltage modification" is issued, and the voice processing module recognizes and replies "Please say, for example, add one volt, or subtract 0.5 volts"; the user gives an instruction according to the example entry, such as issuing "add one volt", and the voice processing module recognizes and replies "Please confirm, reply, confirm, or cancel", and the user gives an instruction according to the example entry, such as issuing "OK", and the voice processing module recognizes and replies "Confirming" and generates a communication instruction and transmits it to the inverter control module. The inverter control module receives the communication, stores the overvoltage value in the storage module, and responds to the voice processing module with the currently modified overvoltage value. The voice processing module receives the communication data and sends an analog signal to the voice broadcast module, such as broadcasting "The current battery overvoltage value is 16.5 volts";

[0087] In terms of timing work, the user issues a voice command of "scheduled power on", the voice processing module recognizes and replies "OK, first tell me the number of hours, such as three hours; then tell me the number of minutes, such as ten minutes"; the user gives instructions according to the example terms, such as "thirty minutes", the voice processing module recognizes and replies "OK", and generates a communication command and transmits it to the inverter control module. After receiving the communication, the inverter control module waits 20 seconds to confirm whether there is other time. If not, it sends a communication command to the voice processing module to confirm the time. After receiving the communication data, the voice processing module sends an analog signal to the voice broadcast module, such as broadcasting "The power on delay of thirty minutes is set."

[0088] The voice processing module uses a neural network intelligent voice chip to perform voice recognition and semantic analysis on electrical signals. The specific implementation method is as follows:

[0089] The digital electrical signal is first framed (20-30ms per frame, with 10-15ms frame overlap) and spectrum leakage is reduced using a Hamming window function. Then, voice endpoint detection (VAD) is performed using a dual-threshold energy detection method combined with zero-crossing rate characteristics to remove silent segments and extract valid speech frames.

[0090] Perform a Fast Fourier Transform (FFT) on valid speech frames to convert time-domain signals into frequency-domain signals, calculate Mel-Frequency Cepstral Coefficients (MFCC) or Linear Prediction Cepstral Coefficients (LPCC), and extract 13-40 dimensional feature vectors to represent the acoustic characteristics of the speech.

[0091] The feature vector is input into a pre-trained neural network model (such as a deep neural network (DNN), a convolutional neural network (CNN), or a recurrent neural network (RNN). Taking the LSTM (Long Short-Term Memory) network-based speech recognition model as an example, the model uses multiple layers of neurons to perform weight calculations and activation functions (such as ReLU) on the feature vector, outputting the probability value of each voice command and selecting the command with the highest probability as the recognition result.

[0092] Establish a vocabulary of inverter professional terms, match keywords in the recognition results using regular expressions, and determine the instruction type;

[0093] Construct grammar rules based on context-free grammar (CFG) and perform grammatical parsing on sentences containing keywords;

[0094] Combining the inverter working status with historical interaction data, the semantic vector representation is optimized through the attention mechanism to eliminate ambiguity and ultimately generate structured semantic instructions.

[0095] Working principle of the inverter circuit: After the inverter main control circuit detects that the battery is connected and recognizes that the battery voltage is normal, it sends a push-pull drive signal to increase the battery voltage to the bus voltage. After detecting that the bus voltage is normal, the inverter main control circuit sends a PWM inverter drive signal to convert the DC carrier into AC power for use by the AC load.

[0096] Working principle of the voice circuit: The microphone converts the language into an electrical signal and enters the language circuit. The voice chip samples through the analog sampling port. The chip has a built-in voice database and recognizes the input voice through comparison. After recognizing the language, if there is a set response voice, it is output to the speaker for broadcast through the chip's digital-to-analog conversion port; the corresponding data can also be sent to other chips through the chip's built-in serial port communication.

[0097] Figure 4 As shown: Sockets 1 and 2: can be connected to AC loads like power strips, such as fans, rice cookers, air conditioners, refrigerators, etc.

[0098] LCD screen: displays output voltage, input voltage, output power, battery capacity, fault information, etc.

[0099] MIC: It is a microphone and a voice receiving device;

[0100] Main switch: Turn on the inverter main switch. Only when this switch is turned on can the inverter have power and work;

[0101] Speaker: Voice output device.

[0102] Voice interaction instructions

[0103] Features: This voice system can not only proactively report faults, but also provide interactive real-time information, the inverter's working status, power on and off, modify the inverter's internal parameters, and repair faults autonomously. It is very flexible and ingenious.

[0104]

[0105]

[0106]

[0107]

[0108] Although the embodiments of the present invention have been shown and described, as detailed above, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the invention, and the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. The inverter voice interaction system is characterized by: include Voice collection and conversion module, which collects the user's voice signal and converts it into an electrical signal; The voice processing module performs voice recognition and semantic analysis on the electrical signal, outputs a voice analog signal based on the semantic analysis results, generates communication instructions as needed, communicates with the inverter control module, and outputs a voice analog signal based on the communication data; The inverter control module collects and processes inverter data, controls the inverter to perform AC / DC conversion, communicates with the voice processing module, and sends communication instructions; The voice broadcast module broadcasts according to the signal of the voice processing module; Storage module, storing the setting parameters of the inverter; Power module, which provides power to the system.

2. The inverter voice interaction system according to claim 1, characterized in that: The voice processing module adopts a neural network intelligent voice chip.

3. The inverter voice interaction system according to claim 1, characterized in that: The working data includes whether the inverter is working, input voltage, output voltage, output power, internal temperature, and fan working status.

4. The inverter voice interaction system according to claim 1, characterized in that: It also includes a display module, which displays the working status and parameters of the inverter.

5. The inverter voice interaction system according to claim 1, characterized in that: Stores the inverter's setting parameters, including over-voltage and under-voltage values, voltage, current, power and electricity calibration values, overload times, overload value, delay time, and historical working data.

6. The inverter voice interaction system according to claim 1, characterized in that: Collect the user's voice signal and convert it into an electrical signal. The specific implementation method is as follows: Use high-sensitivity electret microphone or MEMS microphone; The weak analog electrical signal output by the microphone is amplified by an operational amplifier, and the high-frequency noise is removed by an RC filter circuit, thereby increasing the signal-to-noise ratio to over 60dB. The amplified and filtered analog electrical signal is connected to the ADC, and the analog signal is converted into a digital electrical signal through the sampling theorem; The converted digital electrical signal is temporarily stored in the FIFO buffer built into the FLASH or MCU, waiting to be transmitted to the voice processing module.

7. The inverter voice interaction system according to claim 6, characterized in that: Converted into digital electrical signals through high-performance, low-power audio ADC with SNR ≥ 95dB.

8. The inverter voice interaction system according to claim 1, characterized in that: Generate communication instructions based on the semantic analysis results. The specific implementation method is as follows: According to the inverter control protocol, the structured semantic instructions are mapped into the corresponding communication protocol format; Perform type conversion and encoding on parameters in semantic instructions; Calculate CRC check code or parity bit according to protocol requirements; The mapped protocol data, parameter code and check code are encapsulated into a complete communication frame according to the protocol format and sent to the inverter control module through the communication module.

Images