JP6208610B2 - Fuel cell device - Google Patents

Description

  The present invention relates to a fuel cell device, and more particularly to a fuel cell device that does not make driving sounds uncomfortable.

  In recent years, fuel cell devices have become widespread, and small-scale fuel cell devices are widely used for home use. Since the fuel cell device is operated continuously for a long time, particularly when used at home, the driving sound tends to be an unpleasant noise (noise) mainly at night.

  The fuel cell device increases or decreases the power generation output following the power used by the consumer, and accordingly, the power density-frequency peak value and spectrum of the operating sound fluctuate. It is possible to reduce the power density (energy) at the frequency at which the power density reaches its peak value by taking countermeasures for noise reduction and resonance / resonance prevention. However, there is a limit to the reduction amount of the peak value, and if the peak value is reduced, there is a possibility that the operation sound in other frequency regions that was not noticeable due to the masking effect may appear as an unpleasant sound. It is difficult to eliminate unpleasant noise only by reducing the power density of the frequency. Moreover, even if the overall power density is reduced, the driving sound of a specific frequency still has a peak value, and it is difficult to eliminate unpleasant noise by itself.

  Therefore, it is conceivable to eliminate the unpleasant sound by generating a driving sound that does not feel uncomfortable, instead of eliminating the unpleasant sound by reducing the driving sound. As means for realizing this, there is known one in which a so-called 1 / f fluctuation theory is suitably used (see, for example, Patent Document 1 and Patent Document 2).

  Patent Document 1 describes an air conditioner that varies the amount of blown air with time so as to have a 1 / f fluctuation spectrum. Patent Document 2 describes a vacuum cleaner that rotates the rotation speed of an electric motor in a 1 / f fluctuation pattern.

Japanese Patent Laid-Open No. 3-148557 JP-A-4-189331

  However, in the conventional techniques described in Patent Document 1 and Patent Document 2, although the sound generation source is driven by the 1 / f fluctuation pattern, the actually generated sound is 1 / f. It did not necessarily have a fluctuation spectrum. For this reason, there is a concern that the sound heard by those in the vicinity may become an unpleasant sound, and the generation of the unpleasant sound cannot be sufficiently suppressed.

  The present invention has been invented in view of the above-mentioned conventional problems, and the object of the present invention is to suppress the fluctuation of driving sound from having a spectrum close to 1 / f spectrum and feeling uncomfortable. The object is to provide a fuel cell device.

A fuel cell apparatus according to claim 1 of the present invention is provided.
Output control means for controlling the output;
Measuring means for measuring driving sound;
Based on the time series data of the driving sound measured by the measuring means, spectrum generating means for generating a power spectrum of fluctuation,
Inclination calculating means for calculating the inclination λ of fluctuation that approximately satisfies P = (1 / f) ^λ from the power spectrum of the fluctuation generated by the spectrum generating means, where P is the power density and f is the frequency. When,
Determining means for determining whether or not the inclination λ of the fluctuation calculated by the inclination calculating means is within an allowable range;
When the determination unit determines that the fluctuation slope λ is not within the allowable range, the correction output generation unit generates a correction output in which a predetermined fluctuation is added within the maximum output fluctuation range with the output as a target value. And comprising
The output is controlled by the output control means, and then the driving sound is measured by the measuring means, and then the fluctuation power spectrum is generated by the spectrum generating means, and then the inclination calculating means. The fluctuation slope λ is calculated, and then it is determined by the determination means whether or not the fluctuation slope λ is within the allowable range. When the fluctuation slope λ is not within the allowable range, the correction output generation is performed. The correction output is generated by the means, and feedback control in which the output control by the output control means is performed again based on the correction output is possible.

  According to a second aspect of the present invention, in the first aspect of the invention, the correction output generation unit is configured to output the predetermined fluctuation within the maximum output fluctuation speed and the output fluctuation speed at the output fluctuation speed. It is characterized by setting the time at random.

  According to a third aspect of the present invention, in the first aspect of the invention, the correction output generation unit is configured to cause the predetermined fluctuation to be a sine wave or a triangular wave having a predetermined amplitude and a predetermined period within a maximum output fluctuation speed. It is characterized by being set in a wave shape.

  In the invention according to claim 1 of the present invention, the driving noise actually generated is measured, and when the fluctuation slope λ is not within the allowable range, a correction output is generated, and based on this correction output. Then, feedback control is performed in which the output is controlled again by the output control means. As a result, when the sound source is simply driven so as to have a 1 / f fluctuation spectrum as in the prior art, the fluctuation of the operating sound actually generated has a spectrum close to the 1 / f spectrum. As a result, it is possible to suppress feeling uncomfortable.

  In the invention according to claim 2, it can be expected that the fluctuation slope λ converges within the allowable range quickly by setting fluctuations at random.

  In the invention according to claim 3, since the fluctuation is set to a sine wave or a triangular wave shape that regularly changes, it is not necessary to prepare a plurality of random number patterns, and it is easy to generate a correction output.

It is a block diagram of the fuel cell apparatus of 1st embodiment. It is explanatory drawing explaining installation of the fuel cell apparatus of 1st embodiment. It is a time chart which shows the electric power of a consumer, and the output of a fuel cell apparatus. It is a flowchart of the driving | operation of the fuel cell apparatus of 1st embodiment. (A) and (b) are time charts of fluctuations generated in the first embodiment. (A) and (b) are time charts of fluctuations generated in the second embodiment. (A) and (b) are time charts of fluctuations generated in the third embodiment. It is a time chart of the fluctuation generated in the fourth embodiment.

  A fuel cell device according to a first embodiment of the present invention will be described with reference to FIGS.

  As shown in FIG. 3, the fuel cell device 1 of the present invention includes output control means 2 (see FIG. 1) that increases or decreases the output (generated power) (W) following the power (W) used by the consumer. It is assumed that it is prepared. The fuel cell device 1 is basically continuously operated in a span such as one month, and is not frequently started and stopped. For this reason, when it is used at home, a driving sound is generated at night, and therefore, it is intended to suppress the generated driving sound from being uncomfortable. The fuel cell device 1 of the present invention is not particularly limited to home use, and is not limited to a small-scale device.

  In the present invention, as described in the background art section, the sound pressure (or power density) of the generated driving sound is not reduced, but the fluctuation of the driving sound is uncomfortable based on the so-called 1 / f fluctuation theory. It is intended to be generated so that it does not feel like.

  As shown in FIG. 1, the fuel cell device 1 includes an output control unit 2, a measurement unit 3, a processing unit 4, a spectrum generation unit 5, an inclination calculation unit 6, a determination unit 7, and a correction output generation unit. 8.

  The output control means 2 is a function that the fuel cell device 1 generally has, and a description thereof will be omitted. The operating sound of the fuel cell device 1 varies according to the variation in the output of the fuel cell device 1.

  The measuring means 3 measures the operation sound of the fuel cell device 1, and a general sound collecting microphone device or the like can be used as appropriate, and is not particularly limited. The measuring means 3 obtains time-series data (analog data) of the sound pressure of the driving sound. The measuring means 3 is appropriately installed at a place where driving sound is desired to be measured. In the first embodiment, as shown in FIG. 2, the main body of the fuel cell device 1 is installed immediately adjacent to the building 9 of the consumer, and the operation sound may be uncomfortable for those who live in the adjacent building 98. There shall be. The measuring means 3 may be provided, for example, in the case of a basket 97 provided at the boundary with the adjacent site 99 in the customer site 90. Thereby, without installing the measuring means 3 outside the customer's site 90, it can be provided at a position as close as possible to a person who may feel uncomfortable. Moreover, the measurement means 3 may be provided in the adjacent building 98 as shown in FIG. Thereby, it is possible to measure substantially the same driving sound as the driving sound actually heard by a person living in the adjacent building 98. The time-series data of the sound pressure measured by the measuring means 3 is transmitted to the main body of the fuel cell device 1 by wire or wireless.

  The processing means 4 processes the time-series data of the sound pressure measured by the measuring means 3 so that it can be processed by the spectrum generating means 5. In the first embodiment, sampling is performed, digitization (A / D conversion) is performed, and the data is appropriately stored in a memory or a storage unit. In the first embodiment, the processing unit 4 includes a control unit composed of a microcomputer, and controls other peripheral devices including the storage unit by a program. Known devices can be used as appropriate, and description thereof is omitted. . In general, a sampling period of 25 (milliseconds) is often used because of the processing load and accuracy, but is preferably about 10 (milliseconds) to 50 (milliseconds), and is outside the above range. But it is not particularly limited.

  The spectrum generation unit 5 generates a fluctuation power spectrum based on the time series data of the driving sound measured by the measurement unit 3. Fluctuations are extracted from the time series data, and a fluctuation power spectrum is generated. The fluctuation power spectrum calculates the relationship of power density P-frequency f from fluctuation time-series data, and is performed by Fourier transform (especially so-called FFT is preferable).

The inclination calculation means 6 calculates the fluctuation inclination λ that approximately satisfies P = (1 / f) ^λ from the fluctuation power spectrum generated by the spectrum generation means 5. Here, although a detailed description of the so-called 1 / f fluctuation theory is omitted, the fluctuation slope λ represents the fluctuation power spectrum when the logarithm of the power density P and the frequency f is taken to represent the fluctuation power spectrum. Is the slope of the straight line. The approximation of the fluctuation power spectrum to a straight line is based on the least square method in the first embodiment, but other methods may be used and are not particularly limited. The fluctuation power spectrum is approximated to a straight line by the method of least squares or the like. Of course, the fluctuation power spectrum may be a straight line instead of approximation.

  In the so-called 1 / f fluctuation theory, when the fluctuation slope λ is 1, the power density P is inversely proportional to the frequency f (that is, P = 1 / f), and humans feel comfortable. In the present invention, feedback control is performed so that the fluctuation slope λ calculated by the slope calculation means 6 falls within the allowable range. The allowable range is determined according to the hearing ability of the person who hears the driving sound, the surrounding situation, and the like. In the first embodiment, the allowable range is 0.8 or more and 1.5 or less, but more preferably 0.9 or more and 1.3 or less.

  The determination means 7 determines whether or not the fluctuation inclination λ calculated by the inclination calculation means 6 is within an allowable range.

  The correction output generation means 8 adds a predetermined fluctuation within the maximum output fluctuation range to the output of the fuel cell device 1 as the target value when the determination means 7 determines that the fluctuation slope λ is not within the allowable range. The corrected output is generated. The correction output will be described later.

  In the first embodiment, the spectrum generation unit 5, the inclination calculation unit 6, the determination unit 7, the correction output generation unit 8 and the like are each provided as a part of the function of the control unit composed of a microcomputer. The unit performs overall control, but the spectrum generating means 5 and the like may be provided separately.

  Next, feedback control will be described with reference to FIG.

  When feedback control starts, (S1) output control is performed by the output control means 2. Note that the output control by the output control means 2 is performed even when this feedback control is not performed.

  (S2) The driving noise is measured by the measuring means 3. The measurement time of the driving sound is preferably 30 seconds or longer and within 5 minutes, particularly preferably 1 minute, but is not particularly limited, and may be less than 30 seconds or may exceed 5 minutes. In the first embodiment, it is 5 minutes (300 seconds). That is, it is only necessary that the fluctuation power spectrum and the fluctuation gradient λ can be calculated with a predetermined accuracy.

  (S3) The processing means 4 performs processing such as sampling of the sound pressure time series data measured in (S2), A / D conversion, and the like.

  (S4) The spectrum generation means 5 generates a fluctuation power spectrum based on the time series data of the driving sound processed in (S3).

  (S5) The inclination calculation means 6 calculates the fluctuation inclination λ from the fluctuation power spectrum generated in (S4).

  (S6) The determination means 7 determines whether or not the fluctuation slope λ calculated in (S5) is within an allowable range. If the fluctuation slope λ is within the allowable range in (S6), the feedback control is terminated.

  If the fluctuation slope λ is not within the allowable range in (S6), (S7) the correction output generation means 8 generates a correction output and provides feedback. In (S1) again, based on the correction output. The output control means 2 controls the output. This feedback loop is repeated until the fluctuation slope λ falls within the allowable range. Further, after the end, the next feedback control may be started after a predetermined time or immediately.

  As described above, the present invention measures the driving sound actually generated in the fuel cell device 1 and generates a correction output when the fluctuation slope λ is not within the allowable range. The output control means 2 controls the output again to perform feedback control. As a result, when the sound source is driven so as to have a 1 / f fluctuation spectrum as in the prior art, the actually generated sound is actually compared with those not having the 1 / f fluctuation spectrum. The fluctuation of the generated driving sound has a spectrum close to the 1 / f spectrum, and it is possible to suppress feeling uncomfortable.

  Hereinafter, the correction output generated by the correction output generation means 8 will be described. In the first embodiment, random fluctuations are generated as shown in FIG.

In generating the corrected output, the corrected output generating means 8 randomly sets the output fluctuation speed v and the output fluctuation time Δt at the output fluctuation speed v within the maximum output fluctuation speed v _max . The maximum output fluctuation speed v _max is a physical capability determined by the fuel cell device 1.

Further, the maximum fluctuation range ΔW _max is determined with respect to the output (generated power) of the fuel cell device 1. Maximum absolute value of the variation range ΔW _max | ΔW _max | is 20% or less at least 5% of the output. That is, for the output | [Delta] W _{max |} the small proportion of the effect of fluctuation is reduced, also for the output | [Delta] W _{max |} is large proportion of, undesirable from the viewpoint of stable power supply. Therefore, the range of | ΔW _max | is determined, but the value of | ΔW _max | can be set as appropriate.

When the output of the fuel cell device 1 is the minimum output, it is preferable that the maximum fluctuation range ΔW _max is set by one swing of 0 or more and + | ΔW _max | or less. Further, when the output of the fuel cell device 1 is the rated output, the maximum fluctuation range ΔW _max is set to 0 or less and −−ΔW _max |

The output fluctuation speed v is set randomly within the maximum output fluctuation speed _vmax , and the output fluctuation time Δt is also set randomly. Here, the fluctuation width ΔW (= output fluctuation speed v × output fluctuation time Δt) when the output fluctuation speed v and the output fluctuation time Δt are set does not deviate from the range of the maximum fluctuation width ΔW _max (for example, The upper limit of the output fluctuation time Δt is set to the maximum fluctuation width ΔW _max / the output fluctuation speed v).

  As a random number, a plurality of random number patterns are prepared, a fluctuation pattern as shown in FIG. 5A is generated by one random number pattern, and a fluctuation pattern as shown in FIG. 5B is generated by another random number pattern. Is generated. Then, the correction output generation means 8 feeds back the correction output obtained by adding the fluctuation generated in (S7) (see, for example, FIG. 5A), and controls the output based on the correction output in (S1). Is called. When the fluctuation slope λ is not within the allowable range in (S6), a corrected output is generated by adding another fluctuation (see, for example, FIG. 5B) in (S7) and fed back. If the fluctuation slope λ is not within the allowable range in (S6), another correction output is generated, and thereafter feedback is repeated in the same manner.

  As described above, while another correction output is repeatedly generated in (S7), the fluctuation slope λ normally converges within an allowable range in (S6).

  Furthermore, the converged random number pattern or the converged random number pattern obtained by prior experimentation or the like is defined as a “weighted random number pattern”, and random numbers are generated according to this weighted random number pattern from the next time to generate fluctuations and corrected. By generating the output, convergence becomes even easier.

  In this way, by using random numbers and randomly setting fluctuations, it can be expected that the inclination λ of fluctuations quickly converges within an allowable range.

  Next, a second embodiment will be described based on FIG. The second embodiment is different from the first embodiment in the correction output generated by the correction output generation means 8, and the other points are the same as those in the first embodiment, and the description thereof is omitted.

The description of the maximum output fluctuation speed v _max and the maximum fluctuation width ΔW _max is the same as in the first embodiment, and is omitted. Then, it is set to a triangular wave shape having a predetermined amplitude and a predetermined period.

  For example, the correction output generation means 8 feeds back the correction output generated in (S7) and added with the fluctuation (for example, a triangular wave having an amplitude of 10 W and a period of 40 seconds shown in FIG. 6A), and outputs the next (S1 ), The output is controlled based on the corrected output. When the fluctuation slope λ is not within the allowable range in (S6), correction is performed by adding another fluctuation (for example, a triangular wave having an amplitude of 10 W and a period of 100 seconds shown in FIG. 6B) in (S7). After the output is generated and fed back (S6), when the fluctuation slope λ is not within the allowable range, another correction output is generated, and thereafter the feedback is repeated in the same manner.

  As described above, while another correction output is repeatedly generated in (S7), the fluctuation slope λ normally converges within an allowable range in (S6).

  In the case of the second embodiment, since the fluctuation is set in a triangular wave shape that regularly changes, it is not necessary to prepare a plurality of random number patterns, and it is easy to generate a correction output.

  Next, a third embodiment will be described with reference to FIG. The third embodiment is different from the first embodiment in the correction output generated by the correction output generation means 8, and the other points are the same as those in the first embodiment, and the description thereof is omitted.

The description of the maximum output fluctuation speed v _max and the maximum fluctuation width ΔW _max is the same as in the first embodiment, and is omitted. Then, it is set in a sine wave shape having a predetermined amplitude and a predetermined period.

  For example, the correction output generation means 8 feeds back the correction output generated in (S7) and added with the fluctuation (for example, a sine wave having an amplitude of 10 W and a period of 40 seconds shown in FIG. In S1), the output is controlled based on the correction output. When the fluctuation slope λ is not within the allowable range in (S6), another fluctuation (for example, a sine wave having an amplitude of 10 W and a period of 100 seconds shown in FIG. 7B) is generated in (S7). When the fluctuation slope λ is not within the allowable range in (S6) after being fed back, another correction output is generated, and thereafter the feedback is repeated in the same manner.

  As described above, while another correction output is repeatedly generated in (S7), the fluctuation slope λ normally converges within an allowable range in (S6).

  In the case of the third embodiment, since the fluctuation is set in a sine wave shape that regularly changes, it is not necessary to prepare a plurality of random number patterns, and it is easy to generate a correction output.

  Further, as in the fourth embodiment shown in FIG. 8, a trapezoidal shape other than a triangular wave shape or a sine wave shape may be used, or another shape may be used.

DESCRIPTION OF SYMBOLS 1 Fuel cell apparatus 2 Output control means 3 Measurement means 4 Processing means 5 Spectrum generation means 6 Inclination calculation means 7 Judgment means 8 Correction output generation means 9 User's building 90 Customer's site 97 ９８ 98 Adjacent building 99 Adjacent site

Claims (3)

Output control means for controlling the output;
Measuring means for measuring driving sound;
Based on the time series data of the driving sound measured by the measuring means, spectrum generating means for generating a power spectrum of fluctuation,
Inclination calculating means for calculating the inclination λ of fluctuation that approximately satisfies P = (1 / f) ^λ from the power spectrum of the fluctuation generated by the spectrum generating means, where P is the power density and f is the frequency. When,
Determining means for determining whether or not the inclination λ of the fluctuation calculated by the inclination calculating means is within an allowable range;
When the determination unit determines that the fluctuation slope λ is not within the allowable range, the correction output generation unit generates a correction output in which a predetermined fluctuation is added within the maximum output fluctuation range with the output as a target value. And comprising
The output is controlled by the output control means, and then the driving sound is measured by the measuring means, and then the fluctuation power spectrum is generated by the spectrum generating means, and then the inclination calculating means. The fluctuation slope λ is calculated, and then it is determined by the determination means whether or not the fluctuation slope λ is within the allowable range. When the fluctuation slope λ is not within the allowable range, the correction output generation is performed. The fuel cell apparatus is characterized in that feedback control is possible in which the correction output is generated by the means, and the output is controlled again by the output control means based on the correction output.   2. The fuel cell according to claim 1, wherein the corrected output generation unit randomly sets an output fluctuation speed and an output fluctuation time at the output fluctuation speed within a maximum output fluctuation speed for the predetermined fluctuation. apparatus.   2. The fuel cell apparatus according to claim 1, wherein the correction output generation means sets the predetermined fluctuation to a sine wave or a triangular wave having a predetermined amplitude and a predetermined period within a maximum output fluctuation speed.

Images