JPH0754171B2 - Combustion condition diagnostic device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of use] The present invention relates to a process data detecting means for detecting process data indicating a combustion state in a furnace, and an incineration object into the furnace based on the process data. An expert system for diagnosing and outputting whether the in-reactor combustion state is normal or abnormal based on the process data, for an incinerator having a combustion control device including an input amount control means for increasing / decreasing the input amount The present invention relates to a combustion state diagnosis device using the.

[Conventional technology]

Conventionally, as a combustion state diagnosis device of this type, a combustion control device has a good combustion state such that process data indicating a combustion state detected by using sensors arranged at various places in a combustion chamber falls within a set range. In order to maintain the incinerator, the amount of material to be incinerated was controlled to increase or decrease based on the process data.

Then, a diagnostic device represented by an expert system is provided in the incinerator, and an operator manually performs correction control based on an abnormality diagnosis by the diagnostic device, or a combustion control device is directly used by using the output of the diagnostic device. The control was corrected.

[Problems to be Solved by the Invention]

However, in the above-mentioned expert system, a large amount of incinerated matter with a large amount of water accumulates in the combustion chamber, making it extremely difficult to burn, or a "heap of garbage," or an inflammable incinerated article quickly burns in the combustion chamber to extinguish the fire. It was quite difficult to accurately diagnose an abnormal condition such as a "dust out" condition, and there were cases where it was not possible to identify which abnormal condition.

This is because the process data input to the expert system may be very similar between the "heap of dust" state and the "out of dust" state.

For example, the furnace outlet temperature, which is the temperature of the gas generated by combustion measured upstream of the flue, and the amount of steam generated are
Both the state and the "out of dust" state are low, and the area of the combustion flame in the combustion zone obtained by the industrial TV camera is also small.

In such a case, since it is not possible to identify which abnormal state at that time, it is not possible to immediately perform appropriate control by the operator or the combustion state control device, and as a result, the symptom becomes extremely worse and the combustion There was a drawback that the efficiency was reduced.

An object of the present invention is to eliminate the above-mentioned conventional drawbacks.

[Means for Solving the Problems]

To achieve this object, the combustion state diagnosis device according to the present invention is characterized in that the combustion state diagnosis device is "out of dust".
And when both of the contradictory abnormal states such as "heap of garbage" are diagnostically output, after the incineration object is thrown in by the throwing amount control means for a set time in response to "out of dust", or,
The point is that which abnormal state is specified from the process data after the conveyance of the incineration object by the input amount control means is stopped for a set time corresponding to the “heap of dust”.

[Work]

When both of the contaminated abnormal states of "dust out" and "heap of garbage" are output for diagnosis, if the incinerator is put in for a predetermined time to increase the capacity of the incineration in the combustion chamber, the "absence of dust" will be displayed until then. If it was, it would be improved,
For example, if the furnace outlet temperature, which is the process data, rises, the area of the combustion flame in the combustion zone obtained by an industrial TV camera increases, and if the state is "heap of dust", the state may not change. For example, the furnace outlet temperature, which is the process data, further decreases, and the area of the combustion flame in the combustion zone obtained by the industrial TV camera becomes smaller.

On the contrary, if the incineration of the incinerator is stopped for a predetermined time and the volume of the incinerator in the combustion chamber is reduced, it will be improved if the state of "dust pile" was increased until then. The furnace outlet temperature rises, the area of the combustion flame in the combustion zone obtained by the industrial TV camera increases, and if the state is "dust out", that state does not change or worsens. The furnace outlet temperature, which is the process data, further decreases, and the area of the combustion flame in the combustion zone obtained by the industrial TV camera becomes smaller.

If the above-mentioned predetermined time is set to a tentative time, even if the combustion state deteriorates due to the adverse effect, the diagnosis can be performed while the symptoms are not so serious.

〔The invention's effect〕

Therefore, according to the combustion control device of the incinerator according to the present invention,
Abnormal conditions such as a "heap of garbage" that makes it extremely difficult to burn due to a large amount of incinerated materials that have a large amount of moisture in the combustion chamber, or a "dust-depleted" condition in which inflammable incinerated materials burn quickly in the combustion chamber and extinguish fire. Can be accurately identified and diagnosed in a short period of time, and prompt measures can be taken based on the diagnostic results to improve the condition and improve the combustion efficiency before the symptoms become extremely worse.

〔Example〕

 Embodiments of the present invention will be described below with reference to the drawings.

As shown in Fig. 3, it is composed of an incinerator (1) for municipal waste, a combustion chamber (2), a hopper (3) for throwing in waste, an ash pit (4) for collecting burned ash, etc. There is.

The dust thrown into the hopper (3) is conveyed to the combustion chamber (2) by the pusher (5). The combustion chamber (2) is
It is composed of three blocks, a dry zone (6), a combustion zone (7), and a post-combustion zone (8), and an upward sliding stoker (2a),
It is sequentially transported to each block by (2b) and (2c).

Combustion air is supplied from the hearth and the furnace wall, and the waste is dried in the drying zone (6) with hot air, the waste is incinerated in the combustion zone (7), and the incineration ash in the post-combustion zone (8). Completely burn.

In the combustion zone (7), dust is initially ignited by the burner (9).

The dust burned out in the post combustion zone (8) falls as ash to the ash pushing device (10) and is conveyed to the ash pit (4) by the ash discharging conveyor (11).

The combustion exhaust gas generated in the combustion chamber (2) is taken out as heat energy in the form of steam by a waste heat boiler (12) and supplied to the outside for use as energy for a generator (13) and the like, while an electric dust collector, etc. Exhaust gas treatment equipment consisting of (14)
To remove soot and harmful gas and exhaust.

As shown in FIG. 1, the combustion control device (20) of the incinerator (1) outputs the outputs of various sensors such as a temperature sensor and a gas sensor which are process data detecting means arranged at various places in the combustion chamber (2). Based on the process data as information, the speed of the pusher (5) is controlled to increase / decrease the amount of dust sent to the combustion chamber (2), and at the same time, the stoker speed, pressure, temperature, air amount, etc. are controlled to be appropriate. Combustion control is performed.

That is, the pusher (5) and the combustion control device (20) constitute an input amount control means for increasing / decreasing the input amount of the incineration material into the furnace based on the process data.

However, there are cases in which "dust piles" that are difficult to burn due to a large amount of dust being loaded in the furnace and "out of dust" where there is only a small amount of dust in the furnace.

For example, proper control may not be performed because the identification of these is very confusing, such as when the furnace outlet temperature, which is important data for determining “heap of garbage” or “out of garbage”, decreases. is there.

In this case, a combustion state diagnosis device is provided to diagnose whether or not the combustion control by the combustion control device (20) is normal, and the operation is performed when it is confirmed that an abnormal state occurs based on the output. A staff member performs a correction operation.

The combustion state diagnostic device outputs a combustion state data of the combustion chamber (2), and a neural network (21) which outputs the combustion state data of the combustion chamber (2) and outputs the combustion tendency data of the combustion chamber (2) with the output of the neural network (21) as an input. It consists of an expert system (30).

More specifically, as shown in FIG. 2, the neural network (21) is composed of an input layer (22) composed of a plurality of neurons (22a) composed of a predetermined unit of memory, and an input layer (22). The intermediate layer (23), which is also composed of a plurality of neurons (23a) for intermediate output by associating the output data of the above with the predetermined weight, and the output from the intermediate layer (23) are associated with the predetermined weight, and the combustion chamber (2 ) Output layer (24) consisting of two neurons (24a) that output "dust out index" and "dust pile index" as combustion state data, and input process data from the intermediate layer to the output layer. It is composed of an arithmetic unit (logical element or microcomputer) that multiplies and adds weights.

In the input layer (22), a combustion zone flame or the like which is image information of an industrial television camera (ITV) as an image pickup means (15) for monitoring the combustion state of dust, pusher speed, stoker speed, furnace outlet temperature Output information of various sensors, process data consisting of control output by the combustion control device (20) from 0
It is a normalization in the range of 100, and the average value for the past 10 minutes is input every 1 minute.

"Dust out index" and "Dust pile index" output from the output layer (24) are represented by numerical values in the range of 0.0 to 1.0,
“Good (0.0 to 0.2)”, “Signs (0.4 to 0.6)”,
It is evaluated as “occurrence (0.7 to 1.0)”.

The predetermined weight used in the neural network (21) is a teacher signal input means (25) for inputting, as a teacher signal, combustion process data in which the "dust-missing index" and the "dust pile index" are "occurrence (1.0)". And learning means (26) for adjusting the output of the neural network (21) to "occur (1.0)" corresponding to them when inputting
Are determined in advance by learning (backpropagation) and stored in the memory.

The expert system (30) includes a storage area (31) for storing the output of the neural network (21) every minute as actual data, and a rule base (32) for storing a production rule for determining a combustion state. It is composed of an inference engine (33) which outputs combustion tendency data showing a state of time-series changes from fact data and production rules.

For example, the condition: Out-of-dust index> 0.7 and the pile-up index of dust <0.3 continues for 2 minutes or more. Conclusion: When out-of-dust occurs, condition: 0.4 <Out-of-garde index <0.7, the out-of-garment index rises and the pile-up index of trash <0.3 Conclusion: Sign of garbage outbreak occurs, condition: Waste outage index> 0.7 and garbage pile index> 0.7 Conclusion: It is executed as if the deterioration of burning of garbage or pile of garbage occurs.

In the above-mentioned third example, both the "dust-missing index" and the "dust-heap index" are "0.7" or more, and it is not possible to specify which one. In this case, based on the output information of the expert system (30), for example, output information via a CRT, printer, voice output device, etc., the operator forces the conveyance amount control means to convey the incineration object for a set time. By performing the above, any one is specified based on the process data after the set time has elapsed.

[Another embodiment]

 Another embodiment of the present invention will be described below.

In the above embodiment, a combination of a neural network and an expert system has been described as the combustion state diagnosis device, but the configuration of the combustion state diagnosis device may be arbitrary, and may be composed of only a neural network, for example.

In the above embodiment, when both the "dust-missing index" and the "dust-heap index" are "0.7" or more, the operator forcibly carries the incineration object by the carrying amount control means for the set time. However, the operator may stop the transportation of the incineration object by the transportation amount control means for a set time.

It should be noted that the set time value is not particularly specified and may be set according to the scale of the incinerator.

Furthermore, the operator may not directly control the transportation of the incineration object by the transportation amount control means, but may directly control the transportation amount control means by an output signal of the diagnostic device.

In the above embodiment, the combustion process data is 0 to 100.
The average value for the past 10 minutes is input every 1 minute, but the time series input of the combustion process data is not limited to this and is arbitrary.

In the above embodiment, the neural network has three layers of the input layer, the intermediate layer, and the output layer, but the present invention is not limited to this and is arbitrary.

Furthermore, the output layer is composed of two neurons that output the "dust index" and "dust pile index" as the combustion state data. The neural network outputs the "dust index". , And the one that outputs the “trash pile index” may be used. In this case, the operation speed of the neural network becomes higher.

[Brief description of drawings]

The drawings show an embodiment of a combustion state diagnosis device for an incinerator according to the present invention. Fig. 1 is an overall block diagram, Fig. 2 is a block diagram, and Fig. 3 is a block diagram of the incinerator. (21) …… Neural network, (30) …… Expert system.

 ─────────────────────────────────────────────────── ─── Continuation of front page (56) References JP-A-60-228813 (JP, A) JP-A-63-240602 (JP, A) JP-A-63-51936 (JP, A)

Claims (1)

[Claims] 1. A process data detecting means for detecting process data indicating a combustion state in a furnace, and an input amount control means for increasing / decreasing an input amount of an incineration object into the furnace based on the process data. A combustion state diagnosis device using an expert system for diagnosing and outputting to a incinerator equipped with a combustion control device whether the combustion state in the furnace is normal or abnormal based on the process data. When the status diagnostic device outputs diagnostic output for both contradictory abnormal conditions such as "out of dust" and "heap of dust", the incineration object is thrown in for a set time in response to "out of dust". After that, or after stopping the conveyance of the incineration object by the input amount control means for a set time corresponding to the “heap of dust”, which abnormal state is specified from the process data Combustion condition diagnostic device.