JP7851817B2 - Method for estimating the combustion cutoff position of a waste incinerator and apparatus for estimating the combustion cutoff position of a waste incinerator - Google Patents

Description

This invention relates to a method and apparatus for estimating the combustion cutoff position of a waste incinerator, which estimates the combustion cutoff position of waste being incinerated on the upper surface of the stoker mechanism based on images of the combustion state captured from the downstream side of the stoker mechanism using an imaging device.

Patent Document 1 proposes a waste incinerator equipped with an imaging device that captures flames generated by waste being incinerated on the upper surface of the stoker mechanism from the downstream side of the stoker mechanism, and an image processing device that extracts flame regions from image data obtained by the imaging device and estimates the downstream position of the extracted flame region as the burnout position.

Japanese Patent Application Publication No. 8-94055

However, because the flame generated by the incinerated waste on the upper surface of the stoker mechanism is constantly fluctuating, if the downstream position of the flame region captured by the imaging device is estimated as the burnout position, as in the waste incinerator described in Patent Document 1, the estimated burnout position will fluctuate frequently. As a result, even if automatic combustion control, such as controlling the waste feeding speed by the stoker mechanism or the supply amount of primary combustion air, is performed based on such an estimated burnout position, there is a risk that the control characteristics will deteriorate, making it difficult to maintain a stable combustion state.

Furthermore, if a lumpy object is present within or in front of the flame region as viewed from the imaging device, some of the flame may be obstructed by the object. As a result, the combustion cutoff position may be detected as shifted further upstream than it actually is. In such cases, continuing automatic combustion control could increase the rate at which the stoker mechanism feeds the waste, potentially leading to the discharge of unburned waste.

Therefore, the aforementioned supervisor had to monitor the image showing the incinerator's combustion status displayed on the monitor screen. If they determined that there was a risk of unburned waste being discharged, they had to interrupt the automatic combustion control and manually reduce the waste feeding speed by the stoker mechanism, which required complicated operational intervention.

The object of the present invention, in view of the above-described prior art, is to provide a method and apparatus for estimating the combustion cutoff position of a waste incinerator that minimizes the workload of monitoring personnel and enables continuous automatic operation control.

To achieve the above objective, the first characteristic configuration of the waste incinerator burnout position estimation method according to the present invention is a waste incinerator burnout position estimation method that estimates the burnout position of waste incinerated on the upper surface of the stoker mechanism based on an image of the combustion state taken from the downstream side of the stoker mechanism using an imaging device, wherein the method performs a flame region extraction process that extracts a flame region composed of one or more closed regions by binarizing the image with a predetermined binarization threshold; a burnout position estimation process that estimates the position at the furthest downstream side as the burnout position when a virtual straight line along the furnace width direction is scanned from below to above the flame region, and the ratio of the number of pixels where the virtual straight line and the flame region overlap to the number of pixels of the virtual straight line is a predetermined ratio; an area calculation process that calculates the area of the convex hull flame region obtained by convex hull processing the area of the flame region and the contour of the flame region; and a burnout position adjustment process that adjusts the predetermined ratio based on the relative value of the area of the convex hull flame region to the area of the flame region .

An image containing the flame generated when waste is incinerated on the upper surface of the stoker mechanism is captured by an imaging device from the downstream side of the stoker mechanism. A flame region extraction process is performed on the image captured by the imaging device, and the flame region is extracted by binarizing the pixel values at a predetermined binarization threshold. Next, a burnout position calculation process is performed on the image from which the flame region has been extracted. A virtual straight line along the furnace width direction is scanned from below the flame region upwards, and the downstream position where the ratio of the number of pixels where the virtual straight line and the flame region overlap to the number of pixels in the virtual straight line is a predetermined ratio is estimated as the burnout position. By estimating the burnout position as the position where the number of flame pixels overlapping the virtual straight line at a predetermined ratio downstream of the flame region, it is possible to suppress the drastic fluctuations in the estimated burnout position compared to when the downstream side of the flame region is used as the burnout position.

Furthermore, if flame-retardant or non-combustible lumps are present in the material being incinerated on the upper surface of the stoker mechanism, the flame tends to be obstructed by the lumps, resulting in a smaller flame region extracted by the flame region extraction process. This can make it difficult to distinguish between a decrease in combustion and a reduction in the flame region. Therefore, an area calculation process is performed to calculate the area of the flame region and the area of the convex-hulled flame region obtained by convex-hull processing the outline of the flame region. The area of the convex-hulled flame region includes at least the flame obstructed by the lumps, allowing for an assessment of the degree of influence from the lumps. This is because the area of the convex-hulled flame region when obstructed by lumps tends to be larger than the area of the convex-hulled flame region when combustion decreases and the flame region shrinks. Therefore, by adjusting a predetermined ratio based on the relative value of the area of the convex flame region to the area of the flame region through a combustion cutoff position adjustment process, the risk of misjudging the combustion cutoff position to be upstream of the actual position due to lumpy material can be reduced.

The second characteristic configuration is that, in addition to the first characteristic configuration described above, the fuel cut-off position adjustment process is a process that adjusts the relative value to decrease as the relative value increases when the relative value is greater than a predetermined threshold.

The size of the flame region obstructed by the lumpy material can be evaluated as a relative value. If a predetermined ratio is constant, the larger the relative value, the more likely the combustion cutoff position is to shift upstream. Therefore, assuming that there is a higher risk of misjudging the combustion cutoff position as being upstream of its actual position when the relative value exceeds a predetermined threshold, the predetermined ratio is adjusted to decrease as the relative value increases, thus allowing the system to determine that the combustion cutoff position is downstream.

The third characteristic configuration is that, in addition to the first or second characteristic configuration described above, the relative value is the difference or ratio of the area of the convex hull flame region to the area of the flame region.

It is preferable to use the difference or ratio of the area of the convex hull flame region to the area of the flame region as the relative value of the area of the convex hull flame region to the area of the flame region. When using the difference, it can be assumed that the larger the difference, the higher the risk of misjudging the burn-out position as being upstream of the actual position. Similarly, when using the ratio, it can be assumed that the larger the ratio, the higher the risk of misjudging the burn-out position as being upstream of the actual position.

The fourth characteristic configuration is a method for estimating the burnout position of waste incinerated on the upper surface of a stoker mechanism, which estimates the burnout position of the waste being incinerated on the upper surface of the stoker mechanism based on an image of the combustion state taken from the downstream side of the stoker mechanism using an imaging device, and which includes: a flame region extraction process that extracts a flame region composed of one or more closed regions by binarizing the image with a predetermined binarization threshold; a convex hull process that generates a convex hull flame region, which is the smallest convex polygon that encompasses all the contour points of the flame region; and a burnout position estimation process that estimates the burnout position as the downstreammost position where, when a virtual straight line along the furnace width direction is scanned from below to above the convex hull flame region, the ratio of the number of pixels where the virtual straight line and the convex hull flame region overlap to the number of pixels of the virtual straight line is a predetermined ratio.

When flame-retardant or non-combustible lumps are present in the material being incinerated on the upper surface of the stoker mechanism, the flame tends to be obstructed by these lumps, resulting in a smaller flame region extracted during the flame region extraction process. This can make it difficult to distinguish between this case and a case where the flame region shrinks due to reduced combustion without the presence of lumps. Therefore, by applying a convex hull process, the convex hull flame region, reflecting both the influence of obstruction by lumps and the influence of reduced combustion, is targeted for the burnout position estimation process, allowing for accurate estimation of the burnout position. In the burnout position calculation process, a virtual straight line along the furnace width is scanned from below to above the convex hull flame region. The burnout position is estimated at the downstream end where the ratio of the number of pixels where the virtual line and the convex hull flame region overlap to the number of pixels of the virtual line is a predetermined ratio.

The first characteristic configuration of the waste incinerator burnout position estimation device according to the present invention is a waste incinerator burnout position estimation device that estimates the burnout position of waste incinerated on the upper surface of the stoker mechanism based on an image of the combustion state taken from the downstream side of the stoker mechanism using an imaging device, comprising: a flame region extraction processing unit that extracts a flame region by binarizing the image with a predetermined binarization threshold; a burnout position estimation processing unit that estimates the burnout position as the downstreammost position where the ratio of the number of pixels where the virtual line and the flame region overlap to the number of pixels of the virtual line is a predetermined ratio when a virtual line along the furnace width direction is scanned from below to above the flame region; an area calculation processing unit that calculates the area of the flame region and the area of the convex hull flame region obtained by convex hull processing of the area of the flame region and the contour of the flame region; and a burnout position adjustment processing unit that adjusts the predetermined ratio based on the relative value of the area of the convex hull flame region to the area of the flame region .

The second characteristic configuration is that, in addition to the first characteristic configuration described above, the fuel cut-off position adjustment processing unit adjusts the relative value to decrease as the relative value increases when the relative value is greater than a predetermined threshold.

The third characteristic configuration is that, in addition to the first or second characteristic configuration described above, the relative value is the difference or ratio of the area of the convex hull flame region to the area of the flame region.

The fourth characteristic configuration is a waste incinerator burnout position estimation device that estimates the burnout position of waste incinerated on the upper surface of the stoker mechanism based on an image of the combustion state taken from the downstream side of the stoker mechanism using an imaging device, comprising: a flame region extraction processing unit that binarizes the image with a predetermined binarization threshold to extract a flame region composed of one or more closed regions; a convex hull processing unit that generates a convex hull flame region, which is the smallest convex polygon that encompasses all the contour points of the flame region; and a burnout position estimation processing unit that estimates the burnout position as the furthest downstream position where, when a virtual straight line along the furnace width direction is scanned from below to above the convex hull flame region, the ratio of the number of pixels where the virtual straight line and the convex hull flame region overlap to the number of pixels of the virtual straight line is a predetermined ratio.

As described above, the present invention provides a method and apparatus for estimating the combustion cutoff position of a waste incinerator, which minimizes the workload of monitoring personnel and enables the continuation of automatic operation control.

This is an explanatory diagram of a stoker-type waste incinerator. This is a magnified view of the main components of a stoker-type waste incinerator. (a) and (b) are functional block diagrams of the combustion control device. (a) is an explanatory diagram showing the relative positions of the burning waste on the stoker and the imaging device, and (b) is an explanatory diagram showing the estimation of the burnout position based on the combustion flame captured by the imaging device. (a) is an explanatory diagram illustrating the principle of adjusting the burn-out position based on the area S1 of the combustion flame captured by the imaging device and the area S1' of the flame obtained by convex hull processing of the combustion flame, and (b) is an explanatory diagram illustrating the principle of adjusting the burn-out position based on the area S2 of the combustion flame captured by the imaging device and the area S2' of the flame obtained by convex hull processing of the combustion flame. This diagram illustrates the principle of adjusting the burn-off position based on the flame areas S11' and S12' obtained by hulling each combustion flame, after the combustion flame captured by the imaging device is separated into two areas, S11 and S12, due to the influence of a lumpy object. This is a flowchart showing the combustion control procedure. This flowchart shows the procedure for estimating the fuel cutoff position. This is a flowchart showing the procedure for determining the properties of a substance.

The following describes the method and apparatus for estimating the combustion cutoff position of a waste incinerator according to the present invention, based on the drawings.

[Structure of a waste incinerator]
Figure 1 shows a stoker-type waste incinerator 1. It includes a platform A into which garbage trucks enter, a waste pit B for accumulating the waste collected by the garbage trucks, a waste input hopper D, a waste crane C for transferring waste from waste pit B to waste input hopper D, a furnace chamber E, a waste heat boiler F and economizer G installed in the space above furnace chamber E. The combustion exhaust gas generated in furnace chamber E is purified by exhaust gas treatment equipment such as a cooling tower H and dust collector I arranged along the flue before being exhausted through the chimney J. An induced draft fan L is installed in the flue to maintain negative pressure in furnace chamber E.

By opening the double-hinged waste input door K, located between platform A and waste pit B to prevent odor leakage and ensure safety, the waste collected and transported by the garbage truck is deposited into waste pit B.

The waste collected in waste pit B is grasped by a club bucket type waste crane C, which is operated automatically or by an operator in the control room, and transported to an opening formed at the upper end of the waste input hopper D before being dropped in.

A dust supply device P is installed at the bottom of the waste input hopper D, and the waste filled in the waste input hopper D is pushed into the furnace chamber E. The waste filled in the waste input hopper D functions as a sealing mechanism, blocking the inflow of outside air from the waste input hopper D into the furnace chamber E, thus maintaining a negative pressure in the furnace chamber.

The furnace chamber E comprises a main combustion chamber 2 and a secondary combustion chamber 3 for the complete combustion of the exhaust gas produced in the main combustion chamber 2. Multiple water tubes WT of the waste heat boiler F are embedded in the wall of the secondary combustion chamber 3.

As shown in Figure 2, the main combustion chamber 2 is equipped with a stoker mechanism ST in which fixed and movable grates are alternately arranged along the direction of waste transport. The movable grates are driven back and forth relative to the fixed grates by hydraulic mechanisms h1, h2, and h3, thereby agitating and transporting the waste downstream.

Four airboxes W1, W2, W3, and W4 are installed at the bottom of the stoker mechanism ST, sequentially from upstream to downstream, and the main combustion air is supplied from a forced-air fan. Within the stoker mechanism ST, the upstream region corresponding to airbox W1 is the dry zone ST1, the midstream region corresponding to airboxes W2 and W3 is the combustion zone ST2, and the downstream region corresponding to airbox W4 is the post-combustion zone ST3.

Pressure sensors PS1, PS21, PS22, and PS3 are provided in each of the windboxes W1, W2, W3, and W4, and a pressure sensor PS is provided in the main combustion chamber 2, enabling the detection of the pressure difference between each windbox and the main combustion chamber 2. Furthermore, a flow sensor QS is provided to detect the flow rate of combustion air entering the main combustion chamber 2 via the stoker mechanism ST.

The waste pushed into the main combustion chamber 2 from the dust supply device P is mainly heated and dried in the drying zone ST1, then gasified and burned in the combustion zone ST2. The waste that has been carbonized by gasified combustion is then solid-burned and turned to ash in the post-combustion zone ST3 downstream of the combustion zone ST2 , and after being turned to ash, it falls into the ash chute from the end of the post-combustion zone ST3.

A constricted section is formed in the front wall 2F and rear wall 2R of the furnace chamber E, extending from the main combustion chamber 2 to the inlet of the secondary combustion chamber 3. A gas supply mechanism 4 is provided in this constricted section. The gas supplied from the gas supply mechanism 4 agitates and rectifies the combustion exhaust gas flowing into the secondary combustion chamber 3, allowing for complete combustion in the secondary combustion chamber 3.

The gas supplied from the gas supply mechanism 4 may be air for secondary combustion, exhaust gas drawn from the main combustion chamber 2, recirculated exhaust gas branched from the flue downstream of the dust collector I, or exhaust gas branched from other exhaust gas flow paths, or a mixture of air and the aforementioned exhaust gases.

The total amount of main combustion air and secondary combustion air should be adjusted so that the theoretical air-to-incineration ratio is approximately 1.3. For example, if all the air is supplied by the main combustion air to achieve a theoretical air-to-incineration ratio of approximately 1.3, the gas supplied from the gas supply mechanism 4 may consist only of exhaust gas drawn from the flue. Alternatively, the system may be configured so that approximately 1.0 of the air is supplied by the main combustion air and approximately 0.3 by the secondary combustion air. A temperature sensor and a gas sensor are provided at the outlet of the secondary combustion chamber 3.

An industrial television camera (ITV), acting as an imaging device 5, is installed on the rear wall 2R of the furnace chamber E. The combustion state, including the combustion flame, of the waste being transported and incinerated on the upper surface of the stoker mechanism ST is captured.

[Configuration of combustion control device]
Figure 3 shows the configuration of a combustion control device 10 that controls the combustion state of the waste incinerated in the waste incinerator 1 described above and controls the amount of steam generated in the waste heat boiler F. The combustion control device 10 includes a dust supply control unit 11 that adjusts the amount of waste supplied to the main combustion chamber 2 by a dust supply device P, a transport control unit 12 that controls the transport speed of the dry zone ST1, combustion zone ST2, and post-combustion zone ST3 by hydraulic mechanisms h1, h2, and h3, an air supply control unit 13 that adjusts the amount of main combustion air supplied from each wind box W1 to W4 and also adjusts the amount of air supplied from the gas supply mechanism 4, and a calculation processing unit 14 that outputs control commands to each of the control units 11, 12, and 13.

The arithmetic processing unit 14 includes a combustion cutoff position estimation unit 15 that estimates the combustion cutoff position of waste in the combustion zone ST2, a property determination unit 16 that determines the presence or absence of lumpy material mixed in the waste and its properties such as waste depletion, and a steam amount adjustment unit 17 that adjusts the amount of steam generated by the waste heat boiler F. It also includes a control command generation unit 18 that generates control commands to be output to the control units 11, 12, and 13 according to indicators obtained based on the calculation results of each calculation unit. The detection values of the aforementioned pressure sensors, flow rate sensors, gas sensors, temperature sensors, and steam amount sensors, as well as images captured by the imaging device 5, are input to the arithmetic processing unit 14.

The combustion control device 10 comprises a CPU board, a memory board, an input/output interface board, a display device, an input device, and the like. The combustion control program is installed in the memory on the memory board, and the combustion control program is executed by the CPU on the CPU board, thereby realizing each of the aforementioned functional blocks.

Specifically, as shown in Figure 7, the combustion control device 10 receives values from various sensors such as the pressure sensor, flow sensor, temperature sensor, and gas sensor (SA1), acquires combustion images captured by the imaging device 5 (SA2), and performs combustion cutoff position estimation processing (SA3), property determination processing (SA4), and vapor amount adjustment processing (SA5) based on this input information. Then, based on these results, it performs various control calculations, including PID calculations, to generate control commands to be output to each control unit 11, 12, and 13 (SA6). The dust supply control processing via the dust supply control unit 11 (SA7), the transport control processing via the transport control unit 12 (SA8), and the air supply control processing via the air supply control unit 13 (SA9) are repeatedly executed at predetermined time intervals (SA10). The repetition period of each calculation process is not particularly limited and can be set as appropriate. For example, it is preferable to set the repetition period to several seconds to several tens of seconds.

[Burnout position estimator]
The combustion cutoff position estimation unit 15 is a calculation unit that estimates the combustion cutoff position based on image information showing the combustion state of the waste being incinerated on the upper surface of the stoker mechanism ST, which is captured by the imaging device 5.

The combustion cutoff position estimation unit 15 comprises the following functional blocks: a flame region extraction processing unit 15A, a combustion cutoff position estimation processing unit 15B, an area calculation processing unit 15C, and a combustion cutoff position adjustment processing unit 15D.

As shown in Figure 7, the burnout position estimation unit 15 is configured to repeatedly execute the following processes over time: flame region extraction by the flame region extraction processing unit 15A, burnout position estimation by the burnout position estimation processing unit 15B, area calculation by the area calculation processing unit 15C, and burnout position adjustment by the burnout position adjustment processing unit 15D. That is, it is configured to repeat these processes at predetermined time intervals.

Figure 8 shows the procedure for the combustion cutoff position estimation process performed by the combustion cutoff position estimation unit 15. The flame region extraction processing unit 15A includes a memory for storing moving images captured by the imaging device 5, and a binarization processing unit that extracts flame regions composed of one or more closed regions by binarizing the pixel values of predetermined frame images extracted from the moving images stored in the memory using a predetermined binarization threshold. The image can be composed of any type of pixels, such as a grayscale image or an RGB color image, as long as it allows for the identification of flame regions and other regions. Brightness values, RGB component values, etc., can be appropriately used as the pixel values for extracting flame regions. Furthermore, the binarization threshold is not limited to a specific value; it can be set appropriately as long as it allows for the extraction of flame regions.

The flame region extraction processing unit 15A binarizes the frame image to obtain one or more flame regions (SB1), and distinguishes and extracts a first flame region FA whose size (area) is greater than or equal to a first threshold, and a second flame region FAs whose size (area) is smaller than the first threshold but less than or equal to a second threshold (SB2, SB3). The value of the first threshold should be such that it can extract the largest combustion flame burning on the upper surface of the stoker mechanism, and can extract the main combustion flame even if it is divided into multiple flames due to the influence of a lump-like object described later. The second threshold should be such that it can extract the minute combustion flames excluding the main combustion flame.

The combustion cutoff position estimation processing unit 15B is a functional block that estimates the combustion cutoff position of waste from one or more first flame regions FA whose size (area) is equal to or greater than a first threshold among the extracted flame regions.

In detail, a virtual line VL (a horizontal line in this embodiment) is generated along the furnace width direction for the binarized image. This virtual line VL is then scanned (translated) from below (downstream) to above (upstream) the first flame region FA (SB4). The number of pixels where the virtual line VL and the first flame region FA overlap is calculated (SB5), and the ratio of this number to the total number of pixels on the virtual line VL is calculated. This process is repeated (SB6). The first (farthest downstream) position where the ratio obtained in step SB6 is greater than or equal to a predetermined ratio R is estimated as the burnout position (SB12, SB13).

Considering that the size of the first flame region FA varies depending on whether or not lumpy material is present in the waste incinerated in the combustion zone ST2, the predetermined ratio R is adjusted according to the following procedure.

The adjustment procedure is described below. The area calculation processing unit 15C calculates the area S of the first flame region FA and the area S' of the convex hull flame region obtained by convex hull processing the contour of the first flame region FA (SB7, SB8, SB9). A convex hull is the smallest convex polygon that encompasses all given points, and convex hull processing is the process of generating the smallest convex polygon that encompasses all contour points of the flame image as the convex hull flame region.

The combustion cutoff position adjustment processing unit 15D calculates the relative value RV of the area S' of the convex hull flame region with respect to the area S of the first flame region FA (SB10), and adjusts the predetermined ratio R referenced by the combustion cutoff position estimation processing unit 15B based on the calculated relative value RV (SB11). The relative value RV can be the difference (S'-S) or the ratio (S'/S) between the area S of the first flame region FA and the area S' of the convex hull flame region.

The fuel cut-off position adjustment processing unit 15D adjusts the fuel cut-off position so that, when the relative value RV is greater than a predetermined threshold, the predetermined ratio R decreases as the relative value increases, that is, the fuel cut-off position is corrected to the downstream side. When the relative value RV is less than or equal to the predetermined threshold, the initial value is adopted as the predetermined ratio R. The threshold value of the relative value RV can be determined in advance through tests using an actual machine.

In other words, if flame-retardant or non-combustible lumps are present in the waste incinerated on the upper surface of the stoker mechanism ST, the flame tends to be obstructed by these lumps, resulting in a smaller size of the first flame region FA extracted during the flame region extraction process. This can make it difficult to distinguish between this case and a case where the combustion state deteriorates without the presence of lumps, causing the first flame region FA to shrink.

Therefore, an area calculation process is performed to calculate the area S of the first flame region FA and the area S' of the convex hull flame region obtained by convex hull processing the contour of the first flame region FA. Since the area S' of the convex hull flame region includes at least the flame obstructed by the lumpy material, the degree of influence of the lumpy material can be understood from the area S of the first flame region FA and the area S' of the convex hull flame region. This is because the area S' of the convex hull flame region when obstructed by the lumpy material tends to be larger than the area S' of the convex hull flame region when the combustion state decreases and the first flame region FA becomes smaller.

Therefore, by adjusting a predetermined ratio R based on the relative value of the area S' of the convex hull flame region to the area S of the flame region through a combustion cutoff position adjustment process, the risk of misjudging the combustion cutoff position as being upstream of the actual position due to lumpy material can be reduced.

Figure 4(a) shows an imaging device 5, installed on the rear wall of the furnace chamber E, which photographs the combustion state of waste being incinerated on the upper surface of the stoker mechanism ST from the downstream side of the stoker mechanism ST. In this embodiment, the left-right center of the imaging device 5 is adjusted to the center in the furnace width direction, and the vertical center is adjusted to the center of the vertical length along the transport direction of the drying zone ST1 and combustion zone ST2 within the stoker mechanism ST.

Figure 4(b) shows an image captured by the imaging device 5. The first flame region FA and the second flame regions FAs, generated by gasification combustion in the combustion zone ST2, are hatched. The carbonized waste, gasified and combusted in the combustion zone ST2, is transported to the post-combustion zone ST3, where it is burned to ashes and falls into the ash chute. In Figure 4(b), the upper part of the flame extending above the field of view of the imaging device 5 is approximately flat.

The upper part of Figure 5(a) shows a binarized image of the first flame region FA (the dashed line represents the image lost due to binarization). When a virtual straight line VL (shown as a dashed line) along the furnace width direction is scanned (translated) from below (downstream side of the furnace) to above (upstream side of the furnace) the first flame region FA, the virtual line VL and the first flame region FA come into contact (indicated as "lower end position" in the figure). When scanned (moved) further upward, the virtual line VL and the first flame region FA begin to overlap. The overlapping region is shown as a thick solid line.

The combustion cutoff position estimation processing unit 15B increases the scanning amount (movement amount) of the virtual line VL and estimates the downstream position as the combustion cutoff position when the ratio of the number of pixels PF in the superimposed area of the first flame region FA to the total number of pixels PV of the virtual line VL (PF/PV) becomes a predetermined ratio R.

By estimating the cutoff position in this way, even when the lower end of the first flame region FA frequently fluctuates due to the operation of the stoker mechanism ST and changes in the combustion state, it becomes possible to mitigate the effects and estimate an accurate and stable cutoff position.

The predetermined ratio R is not particularly limited, but in this embodiment, it is set to R = 0.3 (30%) as an initial value. For example, in the case of a video image corresponding to full HD with a horizontal pixel count of 1920, the total number of pixels of the virtual line VL becomes 1920, and the position where the number of pixels PF in the superimposed area of the first flame region FA becomes 1920 × 0.3 = 576 pixels is the burn-out position. In order to reduce the computational load, the system may be configured to calculate the ratio (PF/PV) for an image obtained by downsampling the image at a predetermined downsampling rate.

As shown in the upper part of Figure 5(b), when a lump MT is introduced into the furnace and reaches the combustion zone ST2, and a portion of the combustion flame captured by the imaging device 5 is blocked by the lump MT, a phenomenon occurs where the binarized first flame region FA splits into multiple parts or becomes smaller. As a result, the downstream position where the ratio of the number of pixels PF of the first flame region FA to the total number of pixels PV of the virtual line VL (PF/PV) in the superimposed region with the first flame region FA is a predetermined ratio R shifts upstream compared to the position when the lump MT is not present. In other words, it is incorrectly determined that the combustion cutoff position is upstream. Therefore, the predetermined ratio R is adjusted using the procedure described above.

The lower parts of Figures 5(a) and (b) show the areas S1 and S2 of the hatched flame regions and the areas S1' and S2' of the convex polygons that have undergone convex hull treatment, respectively. When the difference (S' - S) is used as the relative value RV, the relative value RV2 = (S2' - S2) when there is a lump of material MT in the combustion zone ST2 tends to be larger than the relative value RV1 = (S1' - S1) when there is no lump of material MT in the combustion zone ST2.

The fuel cutoff position adjustment processing unit 15D determines that the fuel cutoff position is shifted upstream due to the presence of a lump when the relative value RV = (S1' - S1) is greater than a predetermined threshold. It then adjusts the fuel cutoff position so that the predetermined ratio R decreases as the relative value increases, i.e., so that the fuel cutoff position shifts downstream. If the relative value RV is below the predetermined threshold, it determines that the effect of the lump is small, and the fuel cutoff position is maintained at a pre-set fixed value. The predetermined threshold and adjustment ratio can be determined by conducting tests beforehand to find the correlation between the relative value RV at which the fuel cutoff position is judged to be in the appropriate position and the predetermined ratio, and then determining the correlation based on the found correlation.

Figure 6 shows an example in which the first flame region FA is separated into two by a lump-like object MT. In this case, the downstream position where the ratio (PF/PV) of the total number of pixels PF in the superimposed region of each first flame region FA and the virtual line VL to the total number of pixels PV in the virtual line VL is a predetermined ratio R is estimated to be the burn-off position.

In this case as well, similar to the above, if the relative value RV = {(S11' - S11) + (S12' - S12)} is greater than a predetermined threshold, it is determined that the combustion cutoff position has shifted upstream due to the presence of the lump-like substance MT. Therefore, the predetermined ratio R decreases as the relative value RV increases, meaning the combustion cutoff position is adjusted to move downstream.

Furthermore, if the first flame region FA is divided into two or more regions by a lump-like object MT, the downstream position where the ratio (PF/PV) of the total number of pixels PF of the overlapping region of the first flame region FA and the virtual line VL to the total number of pixels PV of the virtual line VL is a predetermined ratio R may be estimated as the burn-off position for the first flame region FA having the largest area among the multiple first flame regions FA.

[Property determination section]
The properties determination unit 16 is a calculation unit that determines the properties of waste based on image information captured by the imaging device 5 of the waste being incinerated on the upper surface of the stoker mechanism ST.

As shown in Figure 3(b), the property determination unit 16 comprises the following functional blocks: a flame region extraction processing unit 16A, a debris depletion estimation processing unit 16B, an area calculation processing unit 16C, and a clump-like material determination processing unit 16D. The flame region extraction processing unit 16A and the area calculation processing unit 16C are shared functions of the flame region extraction processing unit 15A and the area calculation processing unit 15C that constitute the aforementioned burnout position estimation unit 15.

As shown in Figure 7, similar to the combustion position estimation unit 15, the property determination unit 16 is configured to repeatedly perform the following processes over time, that is, at predetermined time intervals: flame region extraction processing by the flame region extraction processing unit 16A, debris depletion estimation processing by the debris depletion estimation processing unit 16B, area calculation processing by the area calculation processing unit 16C, and clump determination processing by the clump determination processing unit 16D.

Figure 9 shows the procedure for the property determination process performed by the property determination unit 16. As explained in the section on the flame region extraction unit 15A, the flame region extraction unit 16A distinguishes and extracts from the single or multiple flame regions obtained by binarizing the frame image: first flame regions FA whose size (area) is greater than or equal to the first threshold, and second flame regions FAs whose size (area) is smaller than the first threshold but less than or equal to the second threshold (SC1, SC2, SC3).

The area calculation processing unit 16C calculates the area S of the first flame region FA and the area S' of the convex hull flame region obtained by convex hull processing the contour of the first flame region FA (SC4, SC5, SC6). The clump determination processing unit 16D determines whether or not there are clumps of debris MT on the upper surface of the stoker mechanism ST based on the relative value RV of the area S' of the convex hull flame region with respect to the area S of the first flame region FA. As the relative value RV, the difference (S'-S) or ratio (S'/S) of the area S' of the convex hull flame region with respect to the area S of the first flame region FA can be used.

The lump-forming object detection processing unit 16D determines that a lump-forming object MT exists (SC8) when the relative value RV = (S' - S) is greater than a predetermined threshold (SC7, Y), and determines that a lump-forming object MT does not exist (SC9) when the relative value RV = (S' - S) is less than or equal to the predetermined threshold (SC7, N). Furthermore, the lump-forming object detection processing unit 16D determines that the lump-forming object MT is a combustible lump-forming object (SC11) when the relative value RV decreases over time (SC10, Y), and determines that the lump-forming object MT is a non-combustible lump-forming object (SC12) when the relative value RV does not decrease over time (SC10, N).

Furthermore, as shown in Figure 6, if the first flame region FA is separated into two or more parts, the clump determination processing unit 16D determines that a clump MT exists, regardless of the relative value RV = (S' - S), and determines that the clump MT is a combustible clump MT.

The waste depletion estimation processing unit 16B estimates whether there is a tendency for waste depletion based on the combustion cut-off position estimated by the cut-off position estimation processing unit 15B and the second flame regions FAs extracted by the flame region extraction processing unit 16A. Specifically, the waste depletion estimation processing unit 16B estimates a tendency for waste depletion (SC14) when the second flame regions FAs are scattered downstream of the combustion cut-off position (SC13, Y). More specifically, it estimates a tendency for waste depletion when the number of second flame regions FAs located downstream of the combustion cut-off position exceeds a preset waste depletion threshold.

When the thickness of the debris layer upstream of the combustion cutoff point decreases, combustion air supplied from below the stoker mechanism ST blows through, causing debris to be blown downstream of the combustion cutoff point, resulting in the phenomenon of scattered small flame regions downstream of the combustion cutoff point. The debris depletion estimation processing unit 16B can estimate that debris depletion is occurring if the number of these scattered second flame regions downstream of the combustion cutoff point exceeds the debris depletion threshold. The debris depletion threshold can be set appropriately to a value of 2 or greater.

[Combustion control by combustion control device]
The control command generation unit 18 calculates a control value for the transport speed of the stoker mechanism ST based on a predetermined index, and controls the hydraulic mechanisms h1, h2, and h3 via the transport control unit 12 to adjust the transport speeds of the dry zone ST1, combustion zone ST2, and post-combustion zone ST3 that constitute the stoker mechanism ST. In other words, the transport control processing unit is composed of the control command generation unit 18 and the transport control unit 12. The hydraulic mechanisms h1, h2, and h3 are adjusted so that the dry zone ST1, combustion zone ST2, and post-combustion zone ST3 maintain a constant speed ratio while their speeds are controlled.

Furthermore, the transport control processing unit is configured to correct the transport speed of the stoker mechanism ST based on a predetermined index when the lump determination processing unit 16D determines that a lump MT is present. Specifically, if the lump determination processing unit 16D determines that the lump MT is a combustible lump, the transport control processing unit corrects the transport speed by reducing it to ensure sufficient burning time for the lump MT. If the lump determination processing unit 16D determines that the lump MT is a non-combustible lump MT, the transport control processing unit corrects the transport speed by increasing it to quickly discharge the lump MT from the stoker mechanism ST (furnace).

This transport control processing, performed by the transport control processing unit, automatically and appropriately controls the transport speed of the stoker mechanism ST based on pre-set indicators, thereby reducing the need for manual intervention by the operator.

Furthermore, the transport control processing unit is configured to prevent waste depletion from occurring by increasing the transport speed of the stoker mechanism ST based on a predetermined indicator when the waste depletion estimation processing unit 16B estimates that there is a tendency for waste depletion.

As a predetermined indicator, the waste combustion position estimated by the combustion position estimation unit 15 can be used. Specifically, the control command generation unit 18 calculates a control value for the transport speed of the stoker mechanism ST so that the estimated waste combustion position is maintained within a preset control range. At the same time, the control command generation unit 18 calculates a control value for the dust supply device P, and the dust supply speed by the dust supply device P is adjusted via the dust supply control unit 11. At this time, the dust supply speed is increased or decreased in conjunction with the increase or decrease correction of the transport speed of the stoker mechanism ST.

The amount of steam generated by the waste heat boiler F can also be used as a predetermined indicator. The steam amount adjustment unit 17 calculates a target steam amount based on the furnace outlet temperature detected by the temperature sensor and the steam amount detected by the steam amount sensor, so that a predetermined amount of steam is generated. The control command generation unit 18 calculates control values for the combustion air supply amount, the transport speed of the stoker mechanism ST, and the dust supply speed of the dust supply device P, so that the target steam amount is achieved. These control values are input to the air supply control unit 13, the transport control unit 12, and the dust supply control unit 11, and the air supply amount, transport speed, and dust supply speed are controlled accordingly.

Furthermore, if a layer thickness calculation unit is provided in addition to, or separately from, the waste depletion estimation processing unit 16B, to calculate the layer thickness of the waste incinerated by the stoker mechanism ST, it is possible to control the transport speed of the stoker mechanism using the waste layer thickness calculated by the layer thickness calculation unit as an indicator. For example, as a layer thickness calculation unit, an image obtained by binarizing the combustion image of the waste captured by an infrared camera to remove the flames can be obtained, and the average height from the floor surface of the combustion zone ST2 to the surface of the waste can be detected from this image as the waste layer thickness. The specific configuration of the layer thickness calculation unit is not particularly limited, and an existing layer thickness calculation unit configuration can be adopted.

When the aforementioned waste depletion estimation processing unit 16B estimates that there is a tendency for waste depletion, the combustion cutoff position it refers to may be a combustion cutoff position other than the combustion cutoff position estimated by the combustion cutoff position estimation unit 15. For example, the downstream position of the first flame region FA may be adopted as the combustion cutoff position.

In the above-described embodiment, the burnout position estimation unit 15 includes a flame region extraction processing unit 15A that extracts a flame region (first flame region FA ) from a frame image captured by an imaging device, and a virtual straight line along the furnace width direction that scans the flame region (first flame region FA ) from below to above, and the virtual straight line and the flame region (first flame region FA) The configuration described includes a combustion cutoff position estimation processing unit 15B that calculates the position at the downstream end where the ratio of the number of pixels where the two superimposed elements overlap to the number of pixels of the virtual line is a predetermined ratio, an area calculation processing unit 15C that calculates the area of the convex hull flame region obtained by convex hull processing the area of the flame region and the contour of the flame region, and a combustion cutoff position adjustment processing unit 15D that adjusts a predetermined ratio based on the relative value of the area of the convex hull flame region to the area of the flame region. However, the combustion cutoff position estimation unit 15 may be configured with only a flame region extraction processing unit 15A and a combustion cutoff position estimation processing unit 15B, without the area calculation processing unit 15C and the combustion cutoff position adjustment processing unit 15D.

Furthermore, the burnout position estimation unit 15 may be configured to include: a flame region extraction processing unit that binarizes an image at a predetermined binarization threshold to extract flame regions composed of one or more closed regions; a convex hull processing unit that generates a convex hull flame region, which is the smallest convex polygon encompassing all the contour points of the flame region; and a burnout position estimation processing unit that estimates the burnout position as the downstreammost position where, when a virtual line along the furnace width is scanned from below to above the convex hull flame region, the ratio of the number of pixels where the virtual line and the convex hull flame region overlap to the number of pixels of the virtual line is a predetermined ratio.

Furthermore, the embodiments described above are merely examples of the present invention, and it goes without saying that the specific configuration of each part can be appropriately modified and designed within the scope of achieving the effects of the present invention.

1: Waste Incinerator 2: Main Combustion Chamber 3: Secondary Combustion Chamber 4: Gas Supply Mechanism 5: Imaging Device 10: Combustion Control Device 11: Dust Supply Control Unit 12: Conveying Control Unit 13: Air Supply Control Unit 14: Calculation Processing Unit 15: Burnout Position Estimation Unit 15A: Flame Region Extraction Processing Unit 15B: Burnout Position Estimation Processing Unit 15C: Area Calculation Processing Unit 15D: Burnout Position Adjustment Processing Unit 16: Properties Determination Unit 16A: Flame Region Extraction Processing Unit 16B: Waste Depletion Estimation Processing Unit 16C: Area Calculation Processing Unit 16D: Lump-like Object Determination Processing Unit 17: Steam Amount Adjustment Unit 18: Control Command Generation Unit A: Platform B: Waste Pit C: Crane Mechanism D: Waste Input Hopper E: Furnace Chamber F: Waste Heat Boiler G: Economizer FA: Flame Region VL: Virtual Line

Claims (8)

A method for estimating the burnout position of waste being incinerated on the upper surface of a stoker mechanism, which estimates the burnout position of the waste being incinerated on the downstream side of the stoker mechanism based on an image of the combustion state taken from the downstream side of the stoker mechanism using an imaging device,
A flame region extraction process that extracts flame regions composed of one or more closed regions by binarizing the aforementioned image with a predetermined binarization threshold,
A burn-out position estimation process estimates the burn-out position as the position at which the ratio of the number of pixels where the virtual line and the flame region overlap to the number of pixels of the virtual line is a predetermined ratio when a virtual line along the furnace width direction is scanned from below to above the flame region.
Area calculation process to calculate the area of the convex hull flame region obtained by performing convex hull processing on the area of the flame region and the contour of the flame region,
A combustion cutoff position adjustment process that adjusts the predetermined ratio based on the relative value of the area of the convex hull flame region to the area of the flame region,
A method for estimating the burnout position of a waste incinerator. The method for estimating the burn-off position of a waste incinerator according to claim 1 , wherein the burn-off position adjustment process is a process that adjusts the relative value to decrease as the relative value increases when the relative value is greater than a predetermined threshold. The method for estimating the burnout position of a waste incinerator according to claim 1 or 2 , wherein the relative value is the difference or ratio of the area of the convex hull flame region to the area of the flame region. A method for estimating the burnout position of waste being incinerated on the upper surface of a stoker mechanism, which estimates the burnout position of the waste being incinerated on the downstream side of the stoker mechanism based on an image of the combustion state taken from the downstream side of the stoker mechanism using an imaging device,
A flame region extraction process that extracts flame regions composed of one or more closed regions by binarizing the aforementioned image with a predetermined binarization threshold,
A convex hull process that generates a convex hull flame region which is the smallest convex polygon that encompasses all the contour points of the flame region,
A burn-out position estimation process estimates the burn-out position as the position at which the ratio of the number of pixels where the virtual line and the convex flame region overlap to the number of pixels of the virtual line is a predetermined ratio when a virtual line along the furnace width direction is scanned from below to above the convex flame region.
A method for estimating the burnout position of a waste incinerator. A waste incinerator combustion cutoff position estimation device that estimates the combustion cutoff position of waste being incinerated on the upper surface of the stoker mechanism based on an image taken of the combustion state from the downstream side of the stoker mechanism using an imaging device,
A flame region extraction processing unit that binarizes the aforementioned image with a predetermined binarization threshold and extracts the flame region,
A combustion cutoff position estimation processing unit calculates the downstream position as the combustion cutoff position when a virtual straight line along the furnace width direction is scanned from below to above the flame region, and the ratio of the number of pixels where the virtual straight line and the flame region overlap to the number of pixels of the virtual straight line is a predetermined ratio.
An area calculation processing unit that calculates the area of the convex hull flame region obtained by performing convex hull processing on the area of the flame region and the contour of the flame region,
A combustion cutoff position adjustment processing unit adjusts the predetermined ratio based on the relative value of the area of the convex hull flame region to the area of the flame region,
A waste incinerator combustion cutoff position estimation device equipped with the following: The combustion cutoff position estimation device for a waste incinerator according to claim 5 , wherein the combustion cutoff position adjustment processing unit adjusts the predetermined ratio to decrease as the relative value increases when the relative value is greater than a predetermined threshold. The burnout position estimation device for a waste incinerator according to claim 5 or 6 , wherein the relative value is the difference or ratio of the area of the convex hull flame region to the area of the flame region. A waste incinerator combustion cutoff position estimation device that estimates the combustion cutoff position of waste being incinerated on the upper surface of the stoker mechanism based on an image taken of the combustion state from the downstream side of the stoker mechanism using an imaging device,
A flame region extraction processing unit that binarizes the aforementioned image with a predetermined binarization threshold and extracts flame regions consisting of one or more closed regions,
A convex hull processing unit that generates a convex hull flame region which is the smallest convex polygon that encompasses all the contour points of the flame region,
A combustion cut-off position estimation processing unit estimates the downstream position as the combustion cut-off position when a virtual straight line along the furnace width direction is scanned from below to above the convex flame region, and the ratio of the number of pixels where the virtual straight line and the convex flame region overlap to the number of pixels of the virtual straight line is a predetermined ratio.
A waste incinerator combustion cutoff position estimation device equipped with the following: