Method and apparatus for processing signal, controller, cell system, product, and medium

Abstract

Embodiments of the present disclosure relate to a method and apparatus for processing a signal, a controller, a fuel cell system, a computer program product, and a computer-readable storage medium. The method comprises performing, at a first frequency, a plurality of random sub-samplings on an input signal having a second frequency. The method further comprises determining a plurality of target sub-bands in a frequency band of the plurality of random sub-samplings. A target sub-band determined during a single random sub-sampling corresponds to useful information in a sampled signal obtained by the random sub-sampling. The method further comprises combining the target sub-bands determined during the plurality of random sub-samplings into a target frequency band. The method further comprises, on the basis of a sampled signal on the target frequency band, reconstructing the input signal. In the method, an input signal can be sampled and reconstructed.

Description

[0001] Methods, devices, controllers, battery systems, products, and media for processing signals.

[0002] Technical Field

[0003] The embodiments of this disclosure generally relate to the field of signal processing technology, and more specifically to methods, apparatus, controllers, fuel cell systems, computer program products, and computer-readable storage media for processing signals.

[0004] Background Technology

[0005] In some signal processing scenarios, it is necessary to sample analog signals to generate digital signals, and then recover the analog signals from the digital signals on a computing device. In some environments with limited hardware, the sampling frequency may be less than twice the highest frequency of the signal, causing aliasing in the frequency domain, which makes it impossible to recover the analog signal without distortion.

[0006] Summary of the Invention

[0007] Embodiments of this disclosure provide a method, apparatus, controller, fuel cell system, computer program product, and computer-readable storage medium for processing signals.

[0008] According to a first aspect of this disclosure, a method for processing a signal is provided. The method includes performing multiple random subsamplings on an input signal having a second frequency at a first frequency. The method further includes determining multiple target sub-frequency bands within the frequency bands of the multiple random subsamplings. The target sub-frequency bands determined during a single random subsampling process correspond to useful information in the sampled signal obtained in that random subsampling. The method further includes combining the target sub-frequency bands determined during the multiple random subsamplings into a target frequency band. The method further includes reconstructing the input signal based on the sampled signal on the target frequency band.

[0009] According to a second aspect of this disclosure, an apparatus for processing a signal is provided. The apparatus includes a sampling unit, a determining unit, a combining unit, and a reconstruction unit. The sampling unit is configured to perform multiple random sub-samplings on an input signal having a second frequency at a first frequency. The determining unit is configured to determine multiple target sub-frequency bands within the frequency bands of the multiple random sub-samplings. The target sub-frequency bands determined during a single random sub-sampling process correspond to useful information in the sampled signal obtained in that random sub-sampling. The combining unit is configured to combine the target sub-frequency bands determined during the multiple random sub-samplings into a target frequency band. The reconstruction unit is configured to reconstruct the input signal based on the sampled signal on the target frequency band.

[0010] According to a third aspect of this disclosure, a controller is provided. The controller includes at least one processor and a memory. The memory is coupled to the at least one processor and has instructions stored thereon. When executed by the at least one processor, the instructions cause the controller to perform the methods provided according to a first aspect of this disclosure.

[0011] According to a fourth aspect of this disclosure, a fuel cell system is provided. The fuel cell system includes: a controller as described in a third aspect of this disclosure.

[0012] According to a fifth aspect of this disclosure, a computer program product is provided, the computer program product computer-readable medium including computer-executable instructions that, when executed, implement the method provided according to a first aspect of this disclosure.

[0013] According to a sixth aspect of this disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores computer-executable instructions, which are executed by a processor to implement the method provided according to a first aspect of this disclosure.

[0014] Attached Figure Description

[0015] The above and other objects, features and advantages of this disclosure will become more apparent from the accompanying drawings, in which like reference numerals generally denote like parts.

[0016] Figure 1 illustrates a schematic diagram of an example environment in which the apparatus and / or methods according to embodiments of the present disclosure may be implemented.

[0017] Figure 2 is a flowchart illustrating a method for processing signals according to an embodiment of the present disclosure. Figure 3 shows an exemplary schematic diagram for implementing the method shown in Figure 2. Figure 4 shows an exemplary schematic diagram of a process for determining a target sub-frequency band according to an embodiment of the present disclosure.

[0018] Figure 5 shows an exemplary waveform diagram of integrating the sampled signal as shown in Figure 4.

[0019] Figure 6 shows an exemplary schematic diagram of reconstructing a signal using a convex-optimized basis pursuit algorithm.

[0020] Figure 7 shows an exemplary waveform of the input signal.

[0021] Figure 8 shows an exemplary waveform of the reconstructed signal obtained using a convex-optimized basis pursuit algorithm.

[0022] Figure 9 shows a flowchart of reconstructing a signal using the orthogonal matching algorithm in a greedy algorithm.

[0023] Figure 10 shows an exemplary waveform of the reconstructed signal obtained using the orthogonal matching algorithm.

[0024] Figure 11 shows a schematic block diagram of an apparatus for processing signals according to an embodiment of the present disclosure.

[0025] Figure 12 shows a schematic block diagram of a controller suitable for implementing embodiments of the present disclosure.

[0026] In the various accompanying figures, the same or corresponding labels indicate the same or corresponding parts. It should be noted that the elements in the accompanying figures are schematic and not drawn to scale.

[0027] Detailed Implementation

[0028] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0029] In the description of embodiments of this disclosure, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "this embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0030] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this subject matter pertains. It will be further understood that terms such as those defined in commonly used dictionaries shall be interpreted as having the meaning consistent with their meaning in the context of the specification and in the relevant art, and shall not be interpreted in an idealized or overly formal form unless otherwise explicitly defined herein. As used herein, the statement of “connecting” or “coupling” two or more parts together shall mean that these parts are directly joined together or joined through at least one intermediate component.

[0031] >

[0032] In some signal processing scenarios, the sampling frequency achievable by the sampling device may be less than twice the highest frequency of the original signal (in other words, half the sampling frequency may be lower than the highest frequency of the original signal). This prevents the original signal from being reconstructed without distortion, resulting in inaccurate analysis. To accurately analyze the original signal, replacing it with a sampling device that has a higher sampling frequency could be considered. However, replacing it with a higher sampling frequency increases hardware costs. Furthermore, replacing hardware in an existing hardware environment typically requires manual intervention, thus increasing labor costs.

[0033] The aforementioned signal processing scenarios include, for example, the fuel cell system in a fuel cell electric vehicle (FCEV). A fuel cell system is a complex system, its core being the fuel cell stack (also simply referred to as the "stack") based on an electrochemical reaction. In the stack, fuels such as hydrogen and oxygen from the air undergo an electrochemical reaction under the action of a catalyst to generate electrical energy, which serves as the primary power source for the FCEV.

[0034] To monitor the state of a fuel cell stack, electrochemical impedance spectroscopy (EIS) can be used to determine its electrochemical performance. During EIS analysis, a 976 Hz AC signal is injected into the fuel cell stack. The output voltage and current of the fuel cell stack are then sampled, and the impedance is calculated based on these samples. The calculated impedance is then plotted as an impedance spectrum. The horizontal axis of the impedance spectrum represents frequency, and the vertical axis represents impedance amplitude and phase. The shape and characteristics of the impedance spectrum can be used to analyze the electrochemical performance of the fuel cell system. Accurate acquisition of the output voltage and current of the fuel cell stack is crucial for accurate impedance spectrum plotting and thus accurate analysis of the fuel cell system's electrochemical performance.

[0035] In FCEVs, the current sampling module can be reused in multiple application scenarios. The sampling frequency of this current sampling module is typically 1000 Hz. If such a current sampling module is used directly to sample the current output by the fuel cell stack, it will cause aliasing of the sampled signal in the frequency domain, making it impossible to recover the injected 976 Hz AC signal.

[0036] Therefore, embodiments of this disclosure propose a method for processing signals. This method proposes to reconstruct the current signal output by a fuel cell stack using compressed sensing-based analog-to-information (AIC). The method includes performing multiple random sub-samplings on an input signal having a second frequency at a first frequency. The method also includes determining multiple target sub-frequency bands within the frequency bands of the multiple random sub-samplings. The target sub-frequency bands determined during a single random sub-sampling process correspond to useful information in the sampled signal obtained in that random sub-sampling. The method further includes combining the target sub-frequency bands determined during the multiple random sub-samplings into a target frequency band. The method also includes reconstructing the input signal based on the sampled signal on the target frequency band. This method is capable of sampling and reconstructing the input signal. When half of the first frequency is lower than the second frequency, the method can sample and reconstruct the input signal at a sampling frequency lower than twice the highest frequency of the signal, thus making it suitable for hardware environments with lower sampling frequencies.

[0037] Figure 1 illustrates a schematic diagram of an example environment in which the apparatus and / or methods according to embodiments of the present disclosure may be implemented. As shown in Figure 1, a fuel cell system 100 may include a fuel cell stack 101. The fuel cell stack 101 may undergo an electrochemical reaction to generate electrical energy. The fuel cell system 100 may also include a DC / DC converter 105, which can regulate the voltage and current output by the fuel cell stack 101 and convert the varying voltage provided by the fuel cell stack 101 into a stable output voltage. The fuel cell system 100 may also include a fuel cell control unit (FCCU) 104, which can receive signals from and send signals to various components in the fuel cell system 100 to control the various components and achieve overall control of the fuel cell system. In some embodiments, the FCCU 104 can communicate with other components in the fuel cell system 100 via wired or wireless communication methods (including but not limited to controller area network (CAN) bus, local interconnect network (LIN) bus, media oriented systems transport (MOST) bus, vehicle Ethernet, Wi-Fi, Bluetooth, etc.).

[0038] In the example of Figure 1, FCCU 104 can communicate with the DC / DC control unit 110 in the fuel cell system 100. The DC / DC control unit 110 includes, for example, a power transmission unit (PTU) and a power distribution unit (PDU). o During EIS analysis of the fuel cell stack, FCCU 104 can send an instruction to communication module 115 in DC / DC control unit 110 to begin EIS stack detection. Upon receiving this instruction, communication module 115 instructs excitation signal generation module 116 to send a 976 Hz AC signal to DC / DC converter 105. DC / DC converter 105 injects the 976 Hz AC signal into fuel cell stack 101. o

[0039] Voltage sensor 103 acquires the voltage output from fuel cell stack 101 to obtain a voltage signal Uc. Voltage sampling module 112 samples the voltage signal Uc to obtain a sampled voltage signal. Current sensor 102 acquires the current output from fuel cell stack 101 to obtain a current signal Ic. o The current sampling module 111 samples the current signal Ic to obtain a sampled current signal x(t). The sampling frequency of this current sampling module can be less than 976*2 = 1952 Hz, for example, 1000 Hz. Here, the sampled current signal is processed by the signal processing method 200 proposed in the embodiments of this disclosure to obtain a reconstructed sampled current signal X(t). The reconstructed sampled current signal X(t) and the sampled voltage signal output from the voltage sampling module 112 are input to the signal conversion module 113. The signal conversion module 113 converts the time-domain signals of voltage and current into frequency-domain signals respectively by, for example, Fourier transform, and inputs the frequency-domain signals of voltage and current into the impedance calculation module 114. The impedance calculation module 114 calculates the impedance based on the frequency-domain signals of voltage and current and sends the calculated impedance to the communication module 115. The communication module 115 can send the received impedance to the FCCU 104 so that the FCCU 104 can plot an impedance spectrum based on the impedance and analyze the electrochemical performance of the fuel cell system.

[0040] It should be understood that the fuel cell system 100 shown in FIG1 is merely an example of an embodiment of this disclosure and should not be construed as a limitation thereof. For example, in some embodiments, the fuel cell system 100 may include more or fewer components. In the embodiments of this disclosure, the names of the components in the fuel cell system 100 are merely examples; in some embodiments, components having the same or similar functions may have different names. In some embodiments, the components in the fuel cell system 100 may be replaced, and the components before and after replacement may achieve the same or similar functions. It should be understood that the solutions provided in the embodiments of this disclosure can also be applied to other types of fuel cell systems. The fuel cell systems in the embodiments of this disclosure can be applied to various scenarios and can be configured as power sources or auxiliary power in various devices, including but not limited to vehicles, yachts, aerospace equipment, underwater power equipment, etc.

[0041] Figure 2 shows a flowchart of a signal processing method 200 according to an embodiment of the present disclosure. Method 200 can be performed by a signal processing device. This device can be implemented in software and / or hardware, for example, as a controller. The controller can be, for example, a DC / DC control unit 110 in a fuel cell system. The method 200 will now be illustrated schematically using a controller as an example. Referring to Figure 2, method 200 may include blocks 202 to 208. o

[0042] In block 202, the controller performs multiple random subsamplings on the input signal having a second frequency at a first frequency. In this context, the first frequency can also be referred to as the sampling frequency. If the input signal includes multiple frequencies, the second frequency can also be referred to as the highest frequency of the signal. Through random subsampling, the sampled signal is not extended in the frequency domain with a fixed period; the spectrum is no longer neatly shifted, but rather randomly shifted piece by piece. Frequency leakage is evenly distributed throughout the frequency domain, resulting in relatively small leakage values. In block 204, the controller determines multiple target sub-bands within the frequency bands of the multiple random subsamplings. The target sub-bands determined during a single random subsampling process correspond to the useful information in the sampled signal obtained in that random subsampling. In some application scenarios, such as the EIS application scenario described above, the input signal is sparse in the frequency domain, containing only a small number of non-zero values. Therefore, determining the target sub-bands corresponding to the useful information is helpful for subsequent reconstruction operations.

[0043] In block 206, the controller combines the target sub-bands determined during multiple random sub-sampling processes into a target frequency band. This target frequency band indicates which sub-bands contain useful information from the sampled signal.

[0044] In block 208, the controller reconstructs the input signal based on the sampled signal on the target frequency band. Since the target frequency band with useful information has been determined, the input signal can be solved more easily using the useful information.

[0045] The embodiments of this disclosure are capable of sampling and reconstructing input signals, and are particularly suitable for hardware environments with low sampling frequencies. Furthermore, the embodiments of this disclosure only require updating the controller software without changing the hardware, thus saving hardware and labor costs.

[0046] Next, referring to Figure 3, taking a fuel cell system deployed in a vehicle as an example, we will introduce a method for processing signals in some embodiments of this disclosure. In this fuel cell system, the FCCU can be configured to send an indication to start using EIS to detect the fuel cell stack, and the DC / DC control unit can be configured to execute the above-described signal processing method 200. The fuel cell system can be, for example, the aforementioned fuel cell system 100, the FCCU can be, for example, the FCCU 104 in the aforementioned fuel cell system 100, and the DC / DC control unit can be, for example, the DC / DC control unit H0 in the aforementioned fuel cell system 100. o

[0047] After receiving an instruction from FCCU 104 to begin using EIS to detect the fuel cell stack, DC / DC control unit 110 obtains a sampled current signal x(t) at block 310 in Figure 3. The sampled current signal x(t) is obtained by sampling the current signal Ic output by the fuel cell stack in the fuel cell system. The fuel cell stack can be, for example, the fuel cell stack 101 in the aforementioned fuel cell system 100. o The DC / DC control unit 110, at block 310, determines multiple target sub-bands within the frequency range for multiple random sub-samplings of x(t). The sampled signal on a single target sub-band can be represented as y(t). o

[0048] At box 320, the DC / DC control unit 110 stores the target sub-frequency bands determined during multiple random subsampling processes and combines these target sub-frequency bands into a target frequency band. The target sub-frequency bands can be superimposed in the frequency domain to obtain the target frequency band Y(t). Thus, the target frequency band Y(t) includes all target sub-frequency bands y(t).

[0049] At block 330, the DC / DC control unit 110 reconstructs the input signal x(t) based on the sampled signal in the target frequency band to obtain the reconstructed signal X(t). It is desirable that the coupling between X(t) and x(t) is as strong as possible.

[0050] In some embodiments of this disclosure, the time length of the sampled current signal x(t) input to the DC / DC control unit during each random subsampling process is T. T = 1 / fs o fs represents the sampling frequency (1000 Hz in the example of a fuel cell stack). The DC / DC control unit repeatedly samples the current signal Ic with a small bandwidth (less than the sampling frequency) during multiple random sub-sampling processes to obtain the sampled current signal x(t). Assuming the frequency band size for each sampling is Fsub Hz (Fsub < fs), one sampling might be conducted in the band from 0 to Fsub Hz, another in the band from Fsub to 2*Fsub Hz, and so on. After a period of time, sampling can be completed in the band from 0 to fs Hz. Each sampling may identify one or more target sub-bands, or it may determine that no target sub-band exists. Then, the target sub-bands can be combined into a target band, and the sampled signals on the target band are superimposed in the frequency domain. Using a reconstruction method, the reconstructed signal in the band from the target band to fs Hz can be recovered.

[0051] The process of determining the target sub-frequency band in a single random subsampling process (hereinafter referred to as "the current random subsampling") is described below with reference to Figure 4. In Figure 4, the sampled current signal x(t) is a time-domain signal. The time length of the input x(t) in each random subsampling process (hereinafter referred to as "each beat") is T. o T=l / fs ofs represents the sampling frequency (1000 Hz in the example of the fuel cell stack). At block 311, during the current random subsampling process, the input signal is sampled according to the target sub-band determined for the previous random subsampling. The sampled signal S1 on the target sub-band of the current random subsampling can be obtained by the following formula:

[0052] S1 = x(t - l) ⊕ Pc(0, t - l) | x(t) ⊕ Pc(0, t) (1) where ⊕ represents exclusive OR, | represents OR, x(t) represents the sampled current signal obtained by the current random subsampling, x(t - l) represents the sampled current signal obtained by the previous random subsampling, Pc(0, t) is the result of performing a Fourier transform on P(t) during the current random subsampling and then performing a scaling of the power function abstraction range from 3 to 5 from cot(a + θ) to cot(a - θ), a = fs / 2, Pc(0, t - l) is the result of performing a Fourier transform on P(t) during the previous random subsampling and then performing a scaling of the power function abstraction range from 3 to 5 from cot(a + θ) to cot(a - θ). P(t) is a pseudo-random sequence with a period of t. S1 is the form of the result of multiplying the sampled current signal x(t) by the pseudo-random sequence P(t) in the fractional domain, which is actually to shift the signal from t to 0 on the fractional-frequency axis, thereby shifting the signal from the high-frequency band to the low-frequency band. Since the frequency band range with useful information in the current beat has not been determined, and the previous beat is relatively close to the current beat, the useful information distribution will be relatively similar. Therefore, the information of the previous beat is used to generate the sampled signal S1 of the current beat.

[0053] At block 312, x(t) within a first preset time period is integrated, for example, by an integrator, to obtain an integral value 0. Referring to FIG. 5, for example, the integral value 0 can be obtained by the following formula: 0 = C U1 ∫x(t)dt (2) where tmi and tmi+1 represent two consecutive sampling time points, 0 < tm < tmi < tm+1 < T, tm0 = tm, 0 < i < l, m = 1, 2, 3, …, M, tm represents the starting time of sampling, tm+1 represents the starting time of the next sampling. M is a constant. The operation at block 312 is equivalent to performing a conversion from the time domain to the frequency domain.

[0054] At box 313, a comparator is used to compare whether the integral value 0 is within a certain frequency band range [02, 01]. Here, "1" represents the upper limit of the frequency band and "2" represents the lower limit of the frequency band. 01 and 02 can be empirical values. If 0 is within the frequency band range, the comparator outputs 1; if 0 is not within the frequency band range, the comparator outputs 0. In the example of Figure 4, signal S3 is at an active level (e.g., high level) when the comparator output is 1, and signal S4 is at an active level (e.g., high level) when the comparator output is 0.

[0055] At box 314, signal S3 is time-domain encoded using, for example, a D flip-flop to obtain signal S5, while the D flip-flop outputs y(t). y(t) represents useful information in x(t). In the example of Figure 4, signal S3 can be input to the data input D of the D flip-flop, signal S7 can be input to the clock signal CLK of the D flip-flop, the sampled signal y(t) on the target sub-frequency band of the current random subsampling can be obtained from the inverted output Q of the D flip-flop, and signal S5 can be obtained from the non-inverted output Q of the D flip-flop. O

[0056] At box 317, at each first transition edge of signal S4, the counter increments by one. Signal S4 toggles from a second value to a first value at the first transition edge. In one example, the first transition edge is, for example, a rising edge. The first value is, for example, 1, and the second value is, for example, 0. When the counter counts to 7, signal S8 toggles to an active level (e.g., high). Here, / can be an empirical value.

[0057] At box 318, signal S8 is delayed for a second preset time period to obtain signal S9. When signal S9 flips to an active level (e.g., high level), the counter is cleared at box 317. By delaying the clearing, the active level of signal S8 can be maintained for a period of time to allow for subsequent operations.

[0058] At box 316, signal S8 is ORed with the sampled signal y(t) on the target sub-band determined by the previous random subsampling to obtain signal S7. Signal S7 serves as the clock signal for the D flip-flop during the current random subsampling process.

[0059] At box 315, perform an OR operation on signal S5 and a preset time domain range TRange to obtain signal Pc(0,t). oHere, the preset time domain range can be determined based on the preset frequency domain range.

[0060]

[0061] For example, it is a dressing table. e (-j / 2)t 2 9) o

[0062] The process shown in Figure 4 continues until y(t) satisfies the minimum sample number, which is determined using an annealing algorithm. The operations in boxes S315-S318 are equivalent to using the annealing algorithm in determining the target sub-frequency band. The annealing algorithm can reduce sampling power consumption and improve sampling accuracy. Since the fuel cell system can be considered as being powered by a single power source, most values ​​in x(t) are positive, and a small portion are negative. With the help of the annealing algorithm, the set of signals that meet the requirements can be determined through efficient classification, thereby obtaining the minimum sample number. In addition, the process shown in Figure 4 can autonomously select frequency bands with useful information, so that useful information distributed in both low-frequency and high-frequency bands can be preserved, which is more conducive to the subsequent reconstruction effect.

[0063] The process of obtaining the reconstructed signal X(t) is described below with reference to Figures 6 to 10. In Figures 7, 8, and 10, AMP represents the amplitude of the signal, and T represents time.

[0064] Figure 6 shows a simulation diagram of the input signal obtained using a convex-optimized basis pursuit algorithm. In Figure 6, x1 and x2 correspond to the boundaries of the frequency range, and f(x1, x2) corresponds to the magnitude of useful information within that frequency range. The convex-optimized basis pursuit algorithm shown in Figure 6 utilizes existing sparse matrix bases to transform the traditional basis pursuit form into a two-dimensional form to fit the dimension of the sparse matrix. Figure 7 shows an exemplary waveform of the input signal. Figure 8 shows an exemplary waveform of the reconstructed signal obtained using the convex-optimized basis pursuit algorithm. Comparing Figure 8 and Figure 7, it can be seen that the basis pursuit form proposed in this disclosure can make the frequency of the reconstructed signal close to the frequency of the input signal. The inventors of this disclosure have found that the basis pursuit algorithm has relatively high computational requirements on computing devices; therefore, embodiments of this disclosure further propose using an orthogonal matching algorithm from the greedy algorithm to reconstruct the signal.

[0065] Figure 9 shows a flowchart of reconstructing a signal using the orthogonal matching algorithm in a greedy algorithm. Orthogonal matching pursuit is a greedy iterative recovery algorithm that solves the L1 norm minimization problem (finding the sparsest solution of x) with the constraint y=Ax, where A is the sub-Nyquist sampling matrix and y is Y(t).

[0066] At box 902 in Figure 9, the count value n in the iteration counter is initialized to 0, the residual (error of the nth iteration) rn is initialized to y, the solution vector x is initialized to 0, the index set rn of the column in the dictionary is selected and added to the matrix, and the column set Arn is selected and added to the matrix as shown in the observation matrix.

[0067] At box 904, calculate the inner product of the sub-Nyquist sampling matrix A and the residual rn. Then, select the column of the sub-Nyquist sampling matrix A that is most relevant to the residual vector.

[0068] At box 906, the indexes of the newly selected column, the indexes of the newly selected column and the previous index set, and the previous selected column set are set respectively. The reconstructed signal and residual vector of the output are updated, and finally the energy value S| of X(t) is calculated.

[0069] At box 908, determine whether the energy value of X(t) is greater than or equal to twice the noise value (|S|>2*k, where |S| represents the energy value of X(t) and k represents the noise value). If the energy value of X(t) is less than twice the noise value ("N" at box 908), the process returns to box 904 for the next iteration. If the energy value of X(t) is greater than or equal to twice the noise value ("Y" at box 908), the process proceeds to box 910, and the reconstructed signal is output.

[0070] In some embodiments of this disclosure, a weighted frequency band sampling method can be used to extract energy values. A relatively accurate sum of energy values ​​is calculated by weighting the signal set according to the proportions of different frequency bands. Furthermore, the noise value is obtained by mixing and superimposing the higher harmonics generated by the high-frequency power device under AC excitation with the background noise, thus exhibiting good randomness.

[0071] By iteratively fitting an orthogonal sparse matrix basis, the information is restored to the compressed sensing sampling matrix. Orthogonal matching pursuit is relatively simpler than basis pursuit algorithms; referring to Figure 10, the frequency of the reconstructed signal in the simulation is close to the frequency of the input signal. Although not as accurate as the simulation results shown in Figure 8, orthogonal matching pursuit requires less computational power and can more effectively achieve the sampling and reconstruction requirements.

[0072] Figure 11 shows a block diagram of an apparatus 1100 for processing signals according to some embodiments of the present disclosure. As shown in Figure 11, the apparatus 1100 includes a sampling unit 1102, a determination unit 1104, a combination unit 1106, and a reconstruction unit 1108. o Sampling unit 1102 is configured to perform multiple random sub-samplings on an input signal having a second frequency at a first frequency. Determination unit 1104 is configured to determine multiple target sub-frequency bands within the frequency bands of the multiple random sub-samplings. The target sub-frequency bands determined during a single random sub-sampling process correspond to useful information in the sampled signal obtained in that random sub-sampling. Combining unit 1106 is configured to combine the target sub-frequency bands determined during the multiple random sub-samplings into a target frequency band. Reconstruction unit 1108 is configured to reconstruct the input signal based on the sampled signal on the target frequency band.

[0073] In some embodiments of this disclosure, half of the first frequency is lower than the second frequency. In some embodiments of this disclosure, the sampling unit H02 is also configured to sample the input signal according to the target sub-frequency band determined for the previous random sub-sampling during a single random sub-sampling process.

[0074] In some embodiments of this disclosure, the determining unit 1104 includes an integrator, a comparator, a D flip-flop, and a first OR gate. The integrator is configured to integrate the sampled signal obtained from each random subsampling over a first preset time period to obtain an integral value. The comparator is configured to determine whether the integral value is within a preset frequency domain range. The comparator is further configured to determine the value of a first signal as a first value if the integral value is within the preset frequency domain range. The comparator is further configured to determine the value of the first signal as a second value if the integral value is not within the preset frequency domain range. The D flip-flop is configured to perform time-domain encoding on the first signal to obtain a second signal. The first OR gate is configured to perform an OR operation on the second signal and a preset time domain range to obtain a third signal, wherein the preset time domain range is determined based on the preset frequency domain range, and the third signal is used to indicate a target sub-frequency band.

[0075] In some embodiments of this disclosure, the determining unit 1104 further includes a counter, a delay circuit, and a second OR gate. The counter is configured to increment by one at each first transition edge of the fourth signal, wherein the fourth signal is the inverted signal of the first signal. The fourth signal flips from a second value to a first value at the first transition edge. The counter is also configured to generate a fifth signal if the counter value reaches a target count value. The delay circuit is configured to reset the counter to zero after the counter value reaches the target count value, after a second preset time period. The second OR gate is configured to perform an OR operation on the fifth signal and the sampled signal on the target sub-frequency band determined in the previous random sub-sampling to obtain a sixth signal. The D flip-flop is also configured to obtain the sampled signal on the target sub-frequency band of the random sub-sampling based on the first signal and the sixth signal.

[0076] Figure 12 shows a schematic block diagram of a controller 1200 that can be used to implement embodiments of the present disclosure. For example, the DC / DC control unit 110 in Figure 1 can be implemented using the controller 1200. As shown in Figure 12, the controller 1200 includes a processor 1201, which can perform various appropriate actions and processes according to computer program instructions loaded into random access memory (RAM) 1203 based on computer program instructions stored in read-only memory (ROM) 1202. Various programs and data required for the operation of the controller 1200 may also be stored in RAM 1203. The processor 1201, ROM 1202, and RAM 1203 are interconnected to each other via bus 1204. Input / output (I / O) interface 1205 is also connected to bus 1204.

[0077] The various processes and procedures described above, such as method 200, may be executed by processor 1201. For example, in some embodiments, method 200 may be implemented as a software program tangibly contained in a computer-readable medium. In some embodiments, part or all of the software program may be loaded and / or installed on controller 1200 via ROM 1202. When the software program is loaded into RAM 1203 and executed by processor 1201, one or more actions of method 200 described above may be performed.

[0078] In summary, the signal processing method according to the embodiments of this disclosure can sample and reconstruct the input signal at a sampling frequency less than twice the highest frequency of the signal, and thus can be applied to hardware environments with low sampling frequencies.

[0079] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload programmable logic devices (CPLDs), and so on.

[0080] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0081] In the context of this disclosure, a computer-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A computer-readable medium can be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing. Furthermore, although the operations are described in a specific order, this should be understood as requiring that such operations be performed in the specific order shown or in sequential order, or requiring that all illustrated operations be performed to achieve the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations.

[0082] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

[0083] Unless otherwise expressly indicated by the context, the singular form of words used herein and in the appended claims includes the plural form, and vice versa. Thus, when referring to the singular, the plural form of the corresponding term is generally included. Where the term “example” is used herein, particularly when it follows a set of terms, the “example” is merely exemplary and illustrative and should not be considered exclusive or pervasive.

[0084] Further aspects and scope of adaptation become apparent from the description provided herein. It should be understood that various aspects of this application may be implemented individually or in combination with at least one other aspect. It should also be understood that the descriptions and specific embodiments herein are for illustrative purposes only and are not intended to limit the scope of this application. Various embodiments of this disclosure have been described above; the foregoing description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles of the embodiments, their practical application, or improvements to the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

Claims 1. A method for processing signals (200), comprising: The input signal (x(t)) having a second frequency is subjected to (202) multiple random subsamplings at a first frequency; (204) Determine multiple target sub-frequency bands in the frequency bands of the multiple random sub-samplings, wherein the target sub-frequency bands determined in the process of a single random sub-sampling correspond to the useful information in the sampled signal obtained in that random sub-sampling; The target sub-bands determined during the multiple random sub-sampling processes are combined (206) into a target frequency band; and (208) The input signal is reconstructed based on the sampled signal on the target frequency band.

2. The method (200) according to claim 1, wherein performing (202) multiple random subsamplings on the input signal having a second frequency at a first frequency comprises: During a single random subsampling process, the input signal is sampled according to the target sub-frequency band determined for the previous random subsampling.

3. The method (200) according to claim 1, wherein determining the plurality of target sub-bands in the frequency bands of the multiple random sub-samplings comprises, for each random sub-sampling: The sampled signal obtained from the random subsampling is integrated within the first preset time period to obtain the integral value (0); Determine whether the integral value (0) is within a preset frequency range; In response to the integral value (0) being within the preset frequency domain range, the value of the first signal (S3) is determined as the first value; In response to the integral value (0) being outside the preset frequency range, the value of the first signal (S3) is determined as the second value; The first signal (S3) is time-domain encoded to obtain the second signal (S5); the second signal (S5) is ORed with a preset time domain range (TRange) to obtain the third signal (Pc(O)), wherein the preset time domain range (TRange) is determined according to the preset frequency domain range, and the third signal (Pc(O)) is used to indicate the target sub-frequency band.

4. The method (200) according to claim 3, further comprising: At each first transition edge of the fourth signal (S4), the counter is incremented by one, wherein the fourth signal (S4) is the inverted signal of the first signal (S3), and the fourth signal (S4) is flipped from the second value to the first value at the first transition edge; In response to the counter value reaching the target count value, a fifth signal is generated (S8); after the counter value reaches the target count value, the counter is reset to zero after a second preset time period; The fifth signal (S8) is ORed with the sampled signal (S11) on the target sub-frequency band determined by the previous random sub-sampling to obtain the sixth signal (S7); and the sampled signal (y(t)) on the target sub-frequency band of this random sub-sampling is obtained based on the first signal (S3) and the sixth signal (S7).

5. The method (200) according to claim 4, wherein obtaining the sampled signal on the target sub-frequency band of the random sub-sampling based on the first signal (S3) and the sixth signal (S7) comprises: Input the first signal (S3) into the data input terminal of the D flip-flop; The sixth signal (S7) is input to the clock signal terminal of the D flip-flop; And obtain the sampled signal on the target sub-band of the random subsampling from the inverted output of the D flip-flop.

6. The method (200) according to claim 4, wherein time-domain encoding of the first signal (S3) to obtain the second signal (S5) comprises: Input the first signal (S3) into the data input terminal of the D flip-flop; The sixth signal (S7) is input to the clock signal terminal of the D flip-flop; And obtain the second signal from the non-inverting output of the D flip-flop (S5). > 7. The method (200) according to claim 1, wherein the duration of a single random subsampling is the reciprocal of the first frequency.

8. The method (200) according to claim 1, wherein reconstructing the input signal based on the sampled signal in the target frequency band comprises: The input signal is reconstructed using a convex optimization algorithm or an orthogonal matching pursuit algorithm based on the sampled signal in the target frequency band.

9. The method (200) according to claim 1, wherein the input signal is a current signal output by a fuel cell stack, and the fuel cell stack is injected with an AC excitation signal having the second frequency.

10. The method (200) according to claim 1, wherein half of the first frequency is lower than the second frequency.

11. A signal processing apparatus (1100), comprising: A sampling unit (1102) is configured to perform multiple random subsamplings on an input signal having a second frequency at a first frequency; The determination unit (1104) is configured to determine the frequency band of the multiple random sub-samplings. The target sub-frequency bands determined in a single random sub-sampling process correspond to the useful information in the sampled signal obtained in that random sub-sampling; Combined unit 1 6, placed>>> (1 0 ) Its configuration is to combine the target sub-frequency bands determined in the process of the multiple random sub-sampling into a target frequency band; as well as A reconstruction unit (1108) is configured to reconstruct the input signal based on the sampled signal on the target frequency band.

12. A controller (1200), comprising: At least one processor; as well as A memory coupled to the at least one processor and having instructions stored thereon, which, when executed by the at least one processor, cause the controller to perform the method (200) according to any one of claims 1-10.

13. A fuel cell system (100), comprising: The controller (1200) as described in claim 12. > > 14. A computer program product comprising computer-executable instructions that, when executed, implement the steps of the method (200) according to any one of claims 1 to 10.

15. A computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are executed by a processor to implement the steps of the method (200) according to any one of claims 1 to 10.

Images