JP2955431B2 - Incinerator combustion control device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a combustion zone provided with a conveying means for conveying a burning incineration object by a relative movement of a fire grate arranged in front and rear, and a gas combustion of the incineration object in the combustion zone. A burn-off position detecting means for detecting an end position; and a combustion control means for increasing and decreasing the conveying speed of the conveying means so that the burn-off position detected by the burn-off position detecting means falls within a set range. The present invention relates to a combustion control device for an incinerator.

[0002]

2. Description of the Related Art Conventionally, a plurality of burn-off position detecting means are arranged along a conveying direction of an incineration object, and a light quantity detecting means for detecting the presence or absence of a flame, And a grate temperature detection means arranged on the grate along the transport direction of the incineration material and configured to detect the temperature of the grate. In these methods, the size of the flame, which indicates the degree of gas combustion, is determined based on the magnitude of the light amount and the temperature, and the position where the magnitude is smaller than a certain threshold is determined as the burn-out position.

[0003]

However, in the case of using the above-described light amount detecting element, the entire light amount within a certain range of the visual field along the transport direction is detected, so that the cost is high along the transport direction. Unless the light amount detecting element is arranged, detailed position information within the range cannot be obtained, and in the case of detecting the furnace temperature, when the temperature sensor fluctuates significantly due to the change in the combustion air amount. And accurate detection is not always possible. In addition, when the grate temperature is detected, the combustion air blows through the gaps in the garbage, and the part becomes abnormally high temperature or is affected by changes in the temperature or volume of the combustion air. Can not. As a result, any of the above burnouts
Even with the position detecting means, the combustion state of the incinerated material in the combustion zone is not properly maintained, and the burn-out position is shifted to the downstream side, and the incinerated material is transported unburned to the post-combustion zone or burned. There was a disadvantage that the cutting position was shifted to the upstream side and the incineration efficiency was reduced. In view of the above-mentioned conventional disadvantages, the object of the present invention is to consider convection in a furnace, fluctuations in the amount of combustion air, and
The point is to realize a burn-off position detecting means capable of always and accurately detecting the burn-off position irrespective of various disturbance factors such as normal combustion, and maintain an appropriate combustion state.

[0004]

In order to achieve this object, a characteristic configuration of a combustion control apparatus for an incinerator according to the present invention is characterized in that a plurality of the burn-out position detecting means are arranged along a conveying direction of an incinerated object and a flame is provided. A plurality of light quantity detecting means for detecting the presence or absence of the incinerator; a plurality of in-furnace temperature detecting means arranged in the vicinity of the combustion zone for detecting a temperature in the furnace; It consists of a grate temperature detecting means for detecting the temperature of the grate, a plurality of lower neural networks for estimating burn-out points individually based on the detected values from them, and an upper neural network for integrating the outputs of the respective lower neural networks. In that the upper neural network has been trained by a teacher signal created by weighting the output of the lower neural network. .

[0005]

The burn-off position detected by the light quantity detection means, the furnace temperature detection means, and the grate temperature detection means depends on the combustion state of the incineration material, the amount of supplied combustion air, and the like. It cannot be separated and identified linearly depending on the state. Therefore, neural networks suitable for solving such a problem are different from each other in the detection principle of the light amount detection means, the furnace temperature detection means, and the grate temperature detection means.
By using for multiple detection means
Estimate points and use the detected values from them as input
The final burnout point by neural network
Is estimated , and the dust transfer speed, the amount of combustion air, and the like are increased or decreased so that the burn-out position becomes an appropriate position based on the result . Here, the top neural network is
Considering the reliability of multiple lower neural networks
Weight the output of the lower neural network
Since learning has been performed using the created teacher signal, estimation accuracy
Can be mutually complemented, and more accurate estimation is possible.
It becomes.

[0006]

According to the present invention, the light amount detecting means, the furnace temperature detecting means, and the grate temperature detecting means are individually used.
Criterion that is problematic in the case of
Pods, convection in the furnace, fluctuations in the amount of combustion air, localized abnormal combustion
Even if there are various disturbance elements such as burning, fewer detecting elements
Since it is possible to provide a burn-out position detecting means capable of accurately and accurately detecting the burn-out position by the number, the combustion state of the incinerator can always be maintained at an appropriate state.
Unina was Tsu.

[0007]

Embodiments will be described below. As shown in FIG. 3, the incinerator for municipal garbage injects the incineration material stored in the hopper 3 into the combustion chamber 2 by the pusher 5 provided at the lower end thereof, and the incinerator in the combustion chamber 2 Ash extruder 1
It is configured to collect in the ash pit 4 by 0. The combustion chamber 2 is provided with a drying zone 6 for drying and heating the dust introduced by the pusher 5 to a temperature near the ignition point, a combustion zone 7 for burning the dried dust, and an ashes for the dust burned in the combustion zone 7. The combustion zone 8 is arranged in a stepwise manner from the upper side to the lower side, and includes a blower 16a for supplying high-temperature combustion air to a bottom portion thereof, a flow path 16b, and a damper 16c provided in the flow path 16b. Combustion air supply means 16 is provided.

On the floors of the drying zone 6, the combustion zone 7, and the post-combustion zone 8, there is provided a transport means S for stirring and transporting the incineration material loaded thereon toward the ash pit 4 downstream. is there. More specifically, the transporting means S is configured such that the grate G is alternately arranged in the front-rear direction on each of a fixed frame and a frame body which is movable relative to the fixed frame. By sliding the movable frame with a hydraulic cylinder, the front and rear grate G is relatively moved in an oblique vertical direction,
The material to be incinerated thereon is stirred and transported. The combustion zone 7 is composed of two regions, a front combustion zone 7A and a rear combustion zone 7B, and refuse as incineration is mainly gas-combusted in the front combustion zone 7A by a combustion control means 18 described later. And the rear combustion zone 7
In order to completely perform solid combustion in the intermediate portion B, the amount of combustion air supplied by the combustion air supply means 16 and the conveyance means S
Is controlled.

The above-described combustion control apparatus for an incinerator includes a burn-out position detecting means D for detecting the end position of the gas combustion of the incinerated material in the combustion zone 7 and a burn-out detected by the burn-off position detecting means D. Combustion control means 18 and the like for increasing and decreasing the conveying speed of the conveying means S so that the cutting position falls within a set range, and increasing and decreasing the supply amount of combustion air by the combustion air supply means 16 are provided. In order to properly adjust and maintain the combustion state of the combustion chamber 2 based on the detected value, a combustion control including a microcomputer for controlling the injection speed of dust by the pusher 5 and performing the above-described control to adjust and maintain the appropriate combustion state. It comprises means 18 and the like.

[0010] For example, if the burn-out position is upstream of the rear combustion zone 7B, the combustion control means 18 increases the transfer speed while maintaining the furnace outlet temperature at a predetermined temperature, and increases the new garbage input amount. If the burn-out position is on the downstream side of the rear combustion zone 7B, the conveying speed is reduced while maintaining the furnace outlet temperature at a predetermined temperature, and the combustion is performed so that solid combustion can be sufficiently performed. For example, the burn-out position is controlled to be within a set range according to the quality and amount of dust such as the water content at that time, such as increasing the air supply amount.

From the combustion gas generated in the combustion chamber 2, heat energy is taken out in the form of steam to be used as energy of a generator 13 by an exhaust heat boiler 12 and supplied to the outside of the plant. 14 removes soot and harmful gases and exhausts.

As shown in FIG. 2, the burn-off position detecting means D includes a plurality of light quantity detecting means D1 comprising photo sensors which are arranged in a plurality in the conveying direction of the incineration object and detect the presence or absence of a flame based on the intensity of received light. A plurality of in-furnace temperature detecting means D arranged on the upper inner wall near the combustion zone 7 for detecting the temperature in the furnace;
2, a plurality of grate temperature detecting means D3 arranged on the grate G along the transport direction of the incineration material and detecting the temperature of the grate G, and the burn-off points are individually determined based on the detected values from these. A plurality of lower neural networks N to be estimated
1, N2, N3 and each lower neural network N
, N2, and N3.

As shown in FIG. 1, the light amount detecting means D1
Are provided at three places: a first half, a middle part, and a second half of the combustion zone 7. The in-furnace temperature detecting means D2 includes a drying zone, a furnace outlet corresponding to the front combustion zone, and a rear combustion zone. And the grate temperature detecting means D3.
Is constructed by embedding thermocouples T as temperature detecting elements at a total of 12 places, three places in the width direction and four places in the transport direction, in the grate G constituting the floor of the combustion zone 7, Each output signal is input to the lower neural networks N1, N2, N3. However, the grate temperature detecting means D3 inputs the average value of three points in the width direction.

The lower neural networks N1, N
2, N3 learns in advance by back propagation using a typical pattern of the burn-off position detected by each of the light amount detecting means D1, the furnace temperature detecting means D2, and the grate temperature detecting means D3 as a teacher signal. It is configured to output a numerical value between 0.0 and 1.0 as to whether the burn-out position is “upper”, “good”, or “lower” with respect to actual input data. ing. For example, "
"Upper" is 1.0, "good" is 0.0, and "lower" is 0.0
If it is, the judgment based on the input data indicates “upper”, and if “upper” is 0.8, “good” is 0.1, and “lower” is 0.0, it is based on the input data. The judgment indicates that it is almost “up”. That is, the numerical values of “upper”, “good”, and “lower” indicate the respective degrees.

The upper neural network N4 is:
"Upper", "good", "lower" which are the individual burn-out position outputs of the lower neural networks N1, N2, N3.
Is a neural network for comprehensive judgment which takes the three outputs as inputs, respectively. The lower neural networks N1, N2, and N3 corresponding to the light amount detecting means D1, the furnace temperature detecting means D2, and the grate temperature detecting means D3 respectively include: Learning is performed by weighting indicating reliability. That is, the lower neural network N1 corresponds to the light amount detecting means D1, the furnace temperature detecting means D2, and the grate temperature detecting means D3.
"1.0" and lower neural network N2 "
0.9 ”, and“ 0.
8 "is weighted, and learning by back propagation is performed on the input teacher signal shown in Table 2 using the value obtained by the conversion formula shown in Expression 1 as the output teacher signal.

[0016]

[Table 1]

[0017]

(Equation 1)

Another embodiment will be described below. Light amount detecting means D1, furnace temperature detecting means D2, grate temperature detecting means D3
The arrangement position and number of are not particularly limited and are arbitrary. Accordingly, the number of neurons in the input layer of the neural network is arbitrary, and the configuration of the hidden layer is not limited. Lower neural networks N1, N2, N3
Is not always limited to this numerical value, and the actual incinerator structure, the position of the light amount detecting means D1, the in-furnace temperature detecting means D2, and the grate temperature detecting means D3, It will be set appropriately depending on the number.

In the claims, reference numerals are provided for convenience of comparison with the drawings, but the present invention is not limited to the configuration shown in the attached drawings.

[Brief description of the drawings]

FIG. 1 is a block diagram of a main part.

FIG. 2 is a configuration diagram of a neural network.

FIG. 3 is an overall configuration diagram of an incinerator.

[Explanation of symbols]

 7 combustion zone 18 combustion control means D burn-off position detecting means D1 light quantity detecting means D2 furnace temperature detecting means D3 grate temperature detecting means G grate S transport means N1, N2, N3 Lower neural network N4 Upper neural network

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-60-196512 (JP, A) JP-A-60-194219 (JP, A) JP-A-4-161710 (JP, A) (58) Investigation Field (Int.Cl. ⁶ , DB name) F23G 5/50 G05B 13/02 G06F 15/18 520

Claims (1)

(57) [Claims] 1. A combustion zone (7) provided with a conveying means (S) for conveying a burning incineration object by a relative movement of a fire grate (G) arranged in front and rear, and a combustion zone in the combustion zone (7). A burn-out position detecting means (D) for detecting an end position of gas combustion of the incinerated material, and the conveying means (S) such that the burn-off position detected by the burn-off position detecting means (D) falls within a set range. Control means for increasing and decreasing the transport speed of the fuel
8) a combustion control device for an incinerator, comprising: a plurality of burn-out position detecting means (D) arranged along the transport direction of the incineration material to detect the presence or absence of a flame; Detecting means (D1), a plurality of in-furnace temperature detecting means (D2) arranged in the vicinity of the combustion zone (7) for detecting the temperature in the furnace, and the grate (G) along the conveying direction of the incineration material. And a plurality of lower neural networks (N1), (N1), (G3) for estimating a burn-out point individually based on detection values from the grate temperature detection means (D3) which are disposed in N2), (N3) and an upper neural network (N4) integrating the outputs of the lower neural networks (N1), (N2), (N3). The upper neural network (N4) The lower neural network A combustion control device for an incinerator that has been learned by a teacher signal created by weighting the outputs of the workpieces (N1), (N2), and (N3).