JP2009203882A - Combustion analysis method, combustion analysis device, and combustion analysis system for power generation engine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problems of preventing abnormal combustion due to load adjustment and providing quick correspondence at occurrence of the abnormal combustion by recognizing a precursor of abnormal combustion from a correlation of an engine load and an in-cylinder maximum pressure of a power generation engine. <P>SOLUTION: A load-in cylinder maximum pressure data indicating a relationship between the power generation load and the in-cylinder maximum pressure in a predetermined time interval is created from an in-cylinder pressure detected value detected by an in-cylinder pressure detector 38, a regression line indicating the correlation of the power generation load and the in-cylinder maximum pressure is determined on the basis of the load-in cylinder maximum pressure data in a plurality of cylinders of a past constant number of days, threshold value lines are set based upon a standard deviation above and below the regression line, and the precursor of abnormal combustion is determined on the basis of an occurrence ratio of the power generation load and the in-cylinder maximum pressure exceeding the threshold value line. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a combustion diagnostic method, a combustion diagnostic device, and a combustion diagnostic system in a power generation engine including a gas engine and a diesel engine, and more particularly to a relationship between a cylinder pressure in a cylinder and a power generation load over time. It is an invention that makes it possible to estimate deterioration of a combustion control component by detecting a sign of abnormal combustion from a change.

  In power generation gas engines and diesel engines, in order to provide a stable power supply, due to pre-ignition due to high-temperature parts in the combustion chamber or surface ignition of deposited unburned carbon, exfoliated carbon, etc., and uneven mixture concentration, etc. Reliable detection of end-gas self-ignition and knocking, and prompt action, and detection of excessive increase or abnormal decrease in the maximum pressure in the combustion chamber, that is, the maximum in-cylinder pressure. It is required to maintain the durability and stability of the engine by taking countermeasures.

As a diagnostic apparatus for detecting and diagnosing the combustion state in the combustion chamber of the engine, a technique such as Patent Document 1 (Japanese Patent Laid-Open No. 2005-9457) has been proposed.
In Patent Document 1, based on the in-cylinder pressure detection value and the engine crank angle detection value in the combustion chamber of the gas engine, the differential pressure between the compression pressure (P _r ) and the reference pressure (P _b ) at the top dead center of the crank angle. ΔP _r , the compression pressure ratio (ΔP _r / ΔP ₀ ) between the differential pressure ΔP _r between the reference in-cylinder pressure (P ₀ ) and the reference pressure (P _b ) during the compression stroke, and the in-cylinder pressure (P) A technique for determining the presence or absence of occurrence of pre-ignition, knocking, misfire or the like using the in-cylinder pressure change rate (ΔP / Δθ) with respect to the crank angle (θ) is shown. Immediately upon detection of knocking or misfire, combustion control for reduced-cylinder operation and load reduction is performed.

  That is, in the technique disclosed in Patent Document 1, the abnormal combustion detection method for determining whether or not the combustion state at the time of explosion combustion is abnormal in order to maintain stable combustion in the combustion chamber, and the abnormal combustion thereof It is a countermeasure technique at the time of occurrence, and it determines the combustion state for each explosion, and it has not been shown until the future abnormal combustion is predicted in advance from the secular change of the combustion state.

On the other hand, Patent Document 2 (Patent No. 3539759) is a technology that can reliably detect a trouble in a combustion system of a gas engine at an early stage and predict a time until a failure occurs to enable appropriate repair work. No. Gazette) has been proposed.
In Patent Document 2, in a gas engine used in a gas cogeneration system, the exhaust pressure is sampled at a predetermined interval to calculate an average exhaust pressure value of each cylinder, and the average exhaust pressure value of each cylinder is calculated. Necessary for the gas engine before the failure is calculated by comparing the standard deviation with the abnormality determination threshold to determine whether there is a failure and calculating the prediction time from the change in the standard deviation until the engine fails. It is possible to perform repairs such as sufficient parts replacement.

JP 2005-9457 A Japanese Patent No. 3539759

  However, in Patent Document 2, the time until a future failure is predicted by looking at the change in the standard deviation calculated from the average exhaust pressure of each cylinder, but the engine load changes depending on the power generation output. Therefore, the exhaust pressure also fluctuates due to the influence, so it is possible to judge whether the failure is judged or predicted from the change state of the exhaust pressure, whether it is a change in the exhaust pressure due to abnormal combustion or a change in the exhaust pressure due to a change in power generation load. However, it is difficult to make an appropriate failure judgment.

  Therefore, the present invention has been made in view of such a background, and estimates the deterioration of the combustion control component by grasping a sign of abnormal combustion from the correlation between the engine load of the power generation engine and the in-cylinder maximum pressure. Thus, it is an object of the present invention to provide a combustion diagnosis method, a combustion diagnosis device, and a combustion diagnosis system for a power generation engine that can prevent abnormal combustion by load adjustment and can quickly respond when abnormal combustion occurs.

  In order to solve the above-mentioned problem, a first invention is directed to a combustion diagnosis method for a generator engine for predicting abnormal combustion in a cylinder based on a detected value of a cylinder pressure detected by a cylinder pressure detector. Creates in-cylinder maximum pressure data representing the relationship between the power generation load and the in-cylinder maximum pressure at a predetermined time interval from the detected internal pressure value, and based on the load in-cylinder maximum pressure data for a plurality of cylinders for a certain number of past days A regression line indicating the correlation between the power generation load and the maximum in-cylinder pressure is obtained, a threshold line based on the standard deviation is set up and down around the regression line, and the generation load exceeding the threshold line and the maximum in-cylinder pressure are generated. It is characterized in that a sign of abnormal combustion is determined based on the rate.

  A second aspect of the invention relates to a combustion diagnosis apparatus for a generator engine that predicts abnormal combustion in a cylinder based on a detected value of a cylinder pressure detected by a cylinder pressure detector. Based on the maximum pressure data generating means for generating the maximum in-cylinder pressure data representing the relationship between the power generation load and the maximum in-cylinder pressure at a predetermined time interval, and the maximum in-cylinder pressure data for a plurality of cylinders for the past certain days A regression line setting means for obtaining a regression line indicating a correlation between the power generation load and the in-cylinder maximum pressure, a threshold line setting means for setting a threshold line based on the standard deviation vertically around the regression line, and the threshold line. Combustion state determination means for determining a sign of abnormal combustion based on the generation rate of the power generation load exceeding the maximum in-cylinder pressure is provided.

According to the combustion diagnostic method and the invention of the diagnostic apparatus for carrying out the method, since the determination is made using the correlation between the power generation load and the maximum cylinder pressure, whether the change in the cylinder pressure is due to abnormal combustion, Whether it is due to the power generation load can be determined, and combustion abnormality can be accurately determined.
In addition, the secular change of the combustion state can be seen from the frequency of occurrence of in-cylinder pressure that deviates from the threshold line based on the threshold line calculated from the regression line calculated from data for a certain number of days in the past. It is possible to reliably and easily see the change in the combustion state.

  Therefore, compared with a countermeasure method based on the determination result of the combustion state for each explosion combustion, in the present invention, it becomes possible to determine in advance a sign of occurrence of abnormality by looking at a long-term tendency of deterioration of the combustion state. Occasionally, a reduced-cylinder operation or an emergency stop is not performed, and the engine state is stabilized by pre-processing such as a reduction in power generation load, thereby preventing an unplanned stop.

  Preferably, the occurrence rate is calculated for each cylinder based on the number of times that exceeds the threshold straight line with respect to the total number of explosions within a predetermined time. According to such a configuration, the occurrence rate is simple and reliable. It can be calculated.

  Further, in the invention of the diagnostic method, preferably, the regression line is set in a coordinate system in which the horizontal axis indicates the power generation load and the vertical axis indicates the in-cylinder pressure, and the occurrence rate exceeds the upper threshold line. The knocking rate is calculated using the knocking rate, and the occurrence tendency of knocking is determined based on the magnitude of the knocking rate. In the invention of the diagnostic apparatus for performing the diagnostic method, preferably, the regression line setting means is Knocking in which the combustion state determination means calculates the knocking rate based on the occurrence rate exceeding the threshold straight line set above the regression line. It is preferable to have a rate calculating means and determine the tendency of knocking occurrence based on the magnitude of the knocking rate calculated by the knocking rate calculating means.

  According to the method invention and the apparatus invention, since the occurrence tendency of knocking is determined based on the threshold line set on the upper side of the regression line, the secular change of the occurrence state of knocking can be reliably and easily seen.

  Further, in the invention of the diagnostic method, preferably, the regression line is set in a coordinate system in which the horizontal axis indicates the power generation load and the vertical axis indicates the in-cylinder pressure, and the occurrence rate falls below the lower threshold line. The misfire rate is calculated, and the occurrence tendency of misfire is determined by the magnitude of the misfire rate. In the invention of the diagnostic apparatus for performing the diagnosis method, preferably, the regression line setting means is on the horizontal axis. A misfire that calculates a misfire rate based on an occurrence rate that falls below a threshold line that is set below the regression line, while obtaining a regression line in a coordinate system that takes the power generation load and takes the cylinder pressure on the vertical axis. It is preferable to have a rate calculation means and determine the misfire occurrence tendency based on the magnitude of the misfire rate calculated by the misfire rate calculation means.

  According to the method invention and the apparatus invention, since the misfire occurrence tendency is determined by the threshold line set below the regression line, the secular change of the misfire occurrence state can be reliably and easily seen.

  Further, the third aspect of the invention is the load in-cylinder maximum pressure data representing the relationship between the generated load and the in-cylinder maximum pressure at a predetermined time interval while accumulating the in-cylinder pressure detection value and the generation load data of the power generation engine. A remote terminal computer for generating an analysis data file by accumulating data, a combustion diagnosis database server for storing the analysis data file transferred from the remote terminal computer, and a plurality of cylinders for a certain number of past days from the combustion diagnosis database server Reading the load in-cylinder maximum pressure data, obtaining a regression line indicating the correlation between the power generation load and the in-cylinder maximum pressure based on the data, and setting a threshold line based on the standard deviation above and below the regression line, A combustion diagnostic device for determining a sign of abnormal combustion based on a generation rate of the power generation load exceeding the threshold straight line and the maximum in-cylinder pressure; A WEB server for storing the diagnostic result of the diagnostic device and a client computer for providing the diagnostic result from the WEB server, the remote terminal computer, the combustion diagnostic database server, the combustion diagnostic device, and the WEB A server and the client computer are connected via a network.

  According to this invention, since the remote terminal computer, the combustion diagnostic database server, the combustion diagnostic device, the WEB server, and the client computer are connected via a network, By setting a reference value of the occurrence rate for determining a sign of abnormal combustion in advance, the determination of a sign of abnormal combustion can be automated.

That is, the in-cylinder pressure detection value and the power generation load data that are automatically sampled from the engine of the power plant installed in the remote place are accumulated, and the relationship between the power generation load and the maximum cylinder pressure at a predetermined time interval is stored. The maximum in-cylinder pressure data to be represented is accumulated in a terminal computer installed in the remote place to create an analysis data file. The analysis data file is transmitted from the terminal computer to the combustion diagnosis database server installed in the management center or the like via the network.
In the combustion diagnostic apparatus, necessary data in the analysis data file is extracted from the combustion diagnostic database server to determine a sign of abnormal combustion.
The determination result is stored in the combustion diagnosis database server and transferred to the WEB server for storage, and the diagnosis result from the WEB server is sent to the client computer via the network by a request from the client or by periodic transmission. The determination result and the alarm can be provided to the client computer, and the client can check the determination result and the alarm and take an appropriate action.

  As described above, since the determination result output and the alarm output can be automated, the determination result of the abnormal combustion state can be regularly checked by the administrator of the power plant, and an appropriate response can be taken to the tendency of the combustion state to deteriorate. it can.

  According to the present invention, a sign of abnormal combustion is detected from the correlation between the engine load of the power generation engine and the in-cylinder maximum pressure, the deterioration of the combustion control component is estimated, and abnormal combustion is prevented by load adjustment, It is possible to provide a combustion diagnosis method, a combustion diagnosis device, and a combustion diagnosis system for a power generation engine that enable a quick response when combustion occurs.

  Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings. However, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not intended to limit the scope of the present invention unless otherwise specified, and are merely illustrative examples. Only.

FIG. 1 is a configuration diagram showing the overall configuration of a combustion diagnostic device 3 for a power generation gas engine 2 according to an embodiment of the present invention.
In FIG. 1, a gas engine 2 is a multi-cylinder, for example, a 12 to 18 cylinder V-type four-cycle gas engine. A piston 6 is fitted in the cylinder 4 so as to be freely slidable and reciprocating. A crankshaft 10 is provided for conversion to rotation via a grod 8.
The gas engine 2 includes a combustion chamber 12 defined between the upper surface of the piston 6 and the inner surface of the cylinder 4, an air supply port 14 and an air supply pipe 16 connected to the combustion chamber 12, and the air supply port 14 is provided with an air supply valve 18 for opening and closing 14, an exhaust port 20 connected to the combustion chamber 12, and an exhaust valve 22 for opening and closing the exhaust port 20.

A gas mixer 24 is installed in the middle of the air supply pipe 16. The fuel gas supplied from the fuel gas pipe 26 through the gas amount adjusting valve 28 and the compressed air supplied from a supercharger (not shown) (not provided with a supercharger). In this case, non-supercharged air) is mixed by the gas mixer 24, and this premixed gas is supplied to the combustion chamber 12 through the supply port 14 and the supply valve 18.
The ignition device 30 is composed of a pilot injection device that injects a pilot fuel 32 such as light oil into an auxiliary chamber (not shown) by an injection nozzle 34 to ignite and combust it, and injects the combustion flame into the combustion chamber 12.

The fuel control device 36 controls the operation of the gas engine 2 by controlling the opening or opening time of the gas amount adjusting valve 28 and controlling the supply of pilot fuel to the injection nozzle 34.
Further, the in-cylinder pressure of each cylinder 4 is measured by the in-cylinder pressure detector 38 and input to the combustion diagnostic device 3.
Further, the combustion diagnosis device 3 receives the detected value of the crank angle of the gas engine 2 detected by the crank angle detector 42, and further generates the power generation output (power generation) of the generator 44 that is the engine load of the gas engine 2. Load) is input from the power generation output sensor 46 to the combustion diagnostic device 3.
The combustion diagnostic apparatus 3 includes a maximum pressure data creation unit 48, a regression line setting unit 50, a threshold line setting unit 52, a combustion state determination unit 54, and a determination result output unit 56. The result output means 56 outputs a determination result and an alarm to the display / alarm unit 58.

Next, predictive control of abnormal combustion by the combustion diagnostic device 3 will be described based on the flowchart shown in FIG.
First, starting from step S1, in step S2, the in-cylinder pressure data from the in-cylinder pressure detector 38 and the power generation load data from the power generation output sensor 46 are accumulated for each explosion cycle. An analysis data file in which load cylinder maximum pressure data representing the relationship between the power generation load and the cylinder maximum pressure is accumulated is created.
This in-cylinder maximum pressure data is obtained from the in-cylinder pressure data and the power generation load data for each explosion cycle by the maximum pressure data creating means 48, and data of the average generated load for 5 minutes and the maximum in-cylinder pressure for 5 minutes. Are created as data for each cylinder.

Next, in step S3, the load in-cylinder maximum pressure data for the past certain days, for example, the past one week, is extracted, and in step S4, the average load of all cylinders is calculated from the load in-cylinder maximum pressure data in all cylinders. And the average in-cylinder maximum pressure are calculated, and in step S5, the regression indicating the correlation between the power generation load and the in-cylinder maximum pressure using the least square method from the average load and average in-cylinder maximum pressure data of all the cylinders. A straight line K is obtained. The regression line K is set in a coordinate system in which the horizontal axis indicates the power generation load and the vertical axis indicates the in-cylinder pressure.
A series of processing from step S3 to step S5 is performed by the regression line setting means 50.

  Next, in step S6, the threshold line setting means 52 sets a threshold line based on the standard deviation on the upper and lower sides of the set regression line K with the regression line K as the center. The upper threshold straight line S1 and the lower threshold straight line S2 are calculated and set with a deviation of, for example, ± 3 MPa (3σ, σ: standard deviation) on the high and low side of the in-cylinder pressure on the horizontal axis. To do. If the in-cylinder maximum pressure exceeding the upper threshold straight line S1 is generated, knocking occurs, and if the in-cylinder maximum pressure lower than the lower threshold straight line S2 is generated, misfire occurs. Is determined to have occurred.

Next, in step S7, the number of times exceeding the threshold straight line S1 and the number of times falling below S2 are tabulated.
For example, the in-cylinder pressure data and the power generation load data for each explosion cycle in a predetermined cylinder accumulated in the analysis data file for one day 10 days ago are the coordinates where the regression line K and the upper threshold line S1 are set. The state plotted in the system is shown in FIG. This example shows a case where data that is offset upward from the regression line K is plotted.
FIG. 3 (a2) shows a state in which in-cylinder pressure data and power generation load data for each explosion cycle are plotted in a coordinate system in which a regression line K and a lower threshold line S2 are set. This example shows a case where data that is offset below the regression line K is plotted.

  Similarly, FIGS. 3 (b1) and (b2) show a state in which in-cylinder pressure data and power generation load data for each explosion cycle in a predetermined cylinder are plotted in the analysis data file for one day five days ago. 3 (c1) and 3 (c2) show a state in which in-cylinder pressure data and power generation load data for each explosion cycle in a predetermined cylinder accumulated in the analysis data file two days ago are plotted. The combustion state of the predetermined cylinder shown in FIGS. 3 (a) to 3 (c) tends to be worsened by exceeding a large number of thresholds so that knocking is indicated in the X region two days before than ten days ago. As shown in the Y region, it is illustrated that there is a tendency to deteriorate by a number below the threshold value.

In step S8, the number of times in units of one hour is aggregated when the number of times exceeding the threshold is totaled, and in step S9, the result of the aggregated number of times is plotted every hour as a count number in units of time in one day. A trend graph is created and displayed on the display / alarm unit 58.
When creating this trend graph, it is divided into a high-pressure trend graph that indicates the number of times the upper (high-pressure side) threshold line is exceeded, and a trend graph that indicates the number of times that the lower (low-pressure side) threshold line is exceeded. And plot separately to create a trend graph.

  FIG. 4A shows an example of a trend graph on the high pressure side. In this trend graph, the horizontal axis represents the passage of time, the vertical axis represents the number of occurrences per hour, and the current situation is displayed from left to right. The latest results are displayed to be added to the right every day, and the display contents are sequentially sent to the left along with the addition. As a result, the history of the number of occurrences up to 6 days ago is always displayed on one screen. You can see it.

The number of occurrences for the lower threshold straight line S2 on the low pressure side is also shown in FIG. The example of FIG. 4B is a trend graph in which the number of counts for the lower threshold straight line S2 is large and a misfire tendency is indicated.
By displaying such a trend graph on the screen, the power plant manager and the client can check the signs of abnormal combustion by looking at the screen.

In order to determine the sign of abnormal combustion from the trend graph displayed on the display / alarm unit 58 in step S9, considerable skill is required. Therefore, while drawing a trend graph, in the next steps S10 to S12, a sign of abnormal combustion is automatically determined.
First, in step 10, the knocking ratio and the misfire ratio are calculated by (number of times exceeding the threshold straight lines S1, S2 / number of explosions within a predetermined time). For example, the number of explosions within a certain period of time may be the number of explosions within one hour. Thus, since the occurrence rate is calculated based on the number of combustions exceeding the threshold for the total number of explosions within a certain time, the knocking ratio and misfire ratio can be calculated easily and reliably.

In step S11, it is determined whether or not the ratio of the calculation results is equal to or greater than a threshold value. If not, the process ends in step S13. If the ratio is equal to or greater than the threshold value, an alarm is issued in step S12 and countermeasures are taken. Is output.
In addition, about the judgment in step S11,12 (part P of FIG. 2) and the alarm output, the calculated knocking ratio and the magnitude of the misfire ratio are divided into the following cases with respect to the preset threshold values: Treatment may be set.

The ratio is 0 <A <B <C <1.
(1) If the ratio <A,
No warning is issued and a continuous operation instruction is output.
(2) If A <ratio <B,
An alarm is issued, and at the time of knocking, an instruction to reduce the gas supply amount to the corresponding cylinder is output according to the ratio value. Further, at the time of misfire, an instruction to increase the gas supply amount to the corresponding cylinder is output according to the ratio value.
(3) If B <ratio <C,
An alarm is issued, and an instruction to reduce the power generation load is output according to the ratio value at both knocking and misfire.
(4) If C <ratio,
In addition to issuing an alarm, the combustion of the corresponding cylinder is stopped both at the time of knocking and at the time of misfire, and an instruction to reduce the cylinder operation is output.

In this way, by setting the processing items according to the values of the knocking ratio and the misfire ratio, an appropriate treatment instruction and warning according to the combustion state are output.
The combustion state determination means 54 performs the counting process in steps S7 and S8, the calculation of the knocking ratio and the misfire ratio in step S10, and the determination of the combustion state based on the ratio in step S11. In addition, the drawing of the trend graph in step S9, the alarm in S12, and the output of countermeasures are performed by the determination result output means 56 and output to the display / alarm unit 58.

  As described above, since the occurrence frequency of knocking, that is, the occurrence tendency is determined by the threshold line S1 set on the upper side of the regression line K, the secular change in the occurrence state of knocking can be reliably and easily seen. Further, since the occurrence frequency of misfire, that is, the tendency of occurrence, is determined by the threshold line S2 set on the lower side of the regression line K, the secular change of the misfire occurrence state can be reliably and easily seen. And the tendency of abnormal combustion can be caught reliably, not based on the experience and feeling of engineers, and deterioration of the combustion control component can be estimated from the secular change of the combustion state.

  Furthermore, since appropriate countermeasure processing and warning are output according to the magnitude of the knocking ratio and misfire ratio calculated based on the threshold straight lines S1 and S2, the skill for reading the tendency of abnormal combustion from the trend graph is also good. Since it is possible to take countermeasures for stabilizing the engine state, it is possible to prevent unplanned reduced-cylinder operation and engine stoppage.

Next, a combustion diagnosis system for a generator engine in a power plant in which the combustion diagnostic device 3 is installed will be described with reference to FIG.
The in-cylinder pressure detection value and power generation load data of the gas engine 2 for power generation are accumulated, and for example, load in-cylinder maximum pressure data representing the relationship between the power generation load and the in-cylinder maximum pressure at intervals of 5 minutes is accumulated. The analysis data file is created in the remote terminal computer 60 at a remote location where the gas engine 2 for power generation is installed, and stored in the remote terminal computer 60.

  The analysis data file is transferred from the remote terminal computer 60 to the management center where the combustion diagnosis database server 62 is installed via the network and stored. Then, the combustion diagnosis device 3 installed in the management center reads out the load cylinder maximum pressure data for a plurality of cylinders for a predetermined number of days from the combustion diagnosis database server 62, and based on the data, the power generation load and the cylinder maximum pressure A regression line showing the correlation with is obtained.

In the combustion diagnostic apparatus 3, threshold straight lines S1 and S2 based on the standard deviation are set on the upper and lower sides around the regression line K as described above, and the power generation load exceeding the threshold straight line and the in-cylinder maximum pressure are set. A sign of abnormal combustion is determined based on the occurrence rate.
The diagnosis result of the combustion diagnosis device 3 is stored in the database 64 of the combustion diagnosis database server, transferred to the WEB server 66 and stored therein, and the diagnosis result is sent from the WEB server 66 to the client by a request from the client or by periodic transmission. The computer 68 is configured to provide a determination result and an alarm via a network.

  In the combustion diagnosis device 3, a series of combustion diagnosis flows described in FIG. 2 are processed each time so that misfire and knocking are automatically monitored at regular intervals, for example, every hour. The determination result and the alarm accompanying it are output.

As described above, since the remote terminal computer 60, the combustion diagnosis database server 62, the combustion diagnosis device 3, the WEB server 66, and the client computer 68 are connected via the network, an abnormality occurs in the combustion diagnosis device 3. By providing in advance a reference value of the occurrence rate for determining a sign of combustion, it is possible to automatically provide determination results on a regular basis.
And the manager of a power plant etc. can confirm the determination result of an abnormal combustion state regularly, and can take an appropriate response | compatibility with respect to the deterioration tendency of a combustion state.

  According to the present invention, a sign of abnormal combustion is detected from the correlation between the engine load of the power generation engine and the in-cylinder maximum pressure, the deterioration of the combustion control component is estimated, and abnormal combustion is prevented by load adjustment, Since a rapid response at the time of occurrence of combustion is possible, it is useful when applied to a combustion diagnostic apparatus and a combustion diagnostic system.

1 is an overall configuration diagram of a combustion diagnosis device for a power generation gas engine according to an embodiment of the present invention. It is a determination control flowchart of the combustion state of a combustion diagnostic apparatus. It is explanatory drawing which shows the regression line, the upper threshold straight line, the lower threshold straight line, and the plot of a combustion state, (a1, a2) shows the state 10 days ago, (b1, b2) shows the state 5 days ago, (c1 , C2) shows the state two days ago. It is the trend graph which showed the tendency of the count number exceeding a threshold straight line, (a) shows the generating tendency exceeding an upper threshold straight line, (b) shows the generating tendency which fell below the lower threshold straight line. It is explanatory drawing which shows the combustion diagnosis system of the engine for generators in a power plant.

Explanation of symbols

2 Gas engine 3 Combustion diagnostic device 10 Crankshaft 38 In-cylinder pressure detector 44 Generator 46 Power generation output sensor 48 Maximum pressure data creation means 50 Regression straight line setting means 52 Threshold straight line setting means 54 Combustion state judgment means 56 Determination result output means 58 Display / alarm unit 60 Remote terminal computer 62 Combustion diagnostic database server 66 WEB server 68 Client computer

Claims (8)

In the combustion diagnosis method for a generator engine for predicting abnormal combustion in a cylinder based on a detected value in a cylinder pressure detected by a cylinder pressure detector,
The in-cylinder maximum pressure data representing the relationship between the power generation load and the in-cylinder maximum pressure at a predetermined time interval is created from the in-cylinder pressure detection value, and the load in-cylinder maximum pressure data for a plurality of cylinders for a certain number of past days is created. A regression line indicating the correlation between the power generation load and the in-cylinder maximum pressure is obtained, and a threshold line based on the standard deviation is set up and down around the regression line, and the power generation load exceeding the threshold line and the in-cylinder maximum pressure are A method for diagnosing combustion of an engine for a generator, wherein a sign of abnormal combustion is determined based on the occurrence rate of the engine.   2. The generator engine according to claim 1, wherein the occurrence rate is calculated for each cylinder based on the number of times exceeding the threshold straight line with respect to the total number of explosions within a predetermined time. Combustion diagnostic method.   The regression line is set in a coordinate system in which the horizontal axis represents the power generation load and the vertical axis represents the in-cylinder pressure, and the knock rate is calculated based on the occurrence rate exceeding the upper threshold line, and the magnitude of the knock rate 3. The combustion diagnosis method for a generator engine according to claim 2, wherein a tendency of occurrence of knocking is determined by the method.   The regression line is set in a coordinate system in which the horizontal axis represents the power generation load and the vertical axis represents the in-cylinder pressure, and the misfire rate is calculated based on the occurrence rate below the lower threshold line, and the magnitude of the misfire rate is large. 3. The method for diagnosing combustion of a generator engine according to claim 2, wherein the tendency of misfire occurrence is determined according to the above. In the combustion diagnosis device for a generator engine for predicting abnormal combustion in a cylinder based on a detected value in a cylinder pressure detected by a cylinder pressure detector,
Maximum pressure data creating means for creating load in-cylinder maximum pressure data representing the relationship between the power generation load and the in-cylinder maximum pressure at a predetermined time interval from the in-cylinder pressure detection value;
A regression line setting means for obtaining a regression line indicating a correlation between the power generation load and the cylinder maximum pressure based on the cylinder load maximum pressure data in a plurality of cylinders for a certain number of days in the past;
Threshold line setting means for setting a threshold line based on the standard deviation up and down around the regression line;
A combustion diagnosis apparatus for a generator engine, comprising combustion state determination means for determining a sign of abnormal combustion based on a generation rate of a power generation load exceeding the threshold straight line and an in-cylinder maximum pressure.   The regression line setting means obtains a regression line in a coordinate system in which the horizontal axis indicates the power generation load and the vertical axis indicates the in-cylinder pressure, and the combustion state determination means generates a value exceeding a threshold line set above the regression line. 6. The generator according to claim 5, further comprising a knocking rate calculating unit that calculates a knocking rate based on a rate, wherein the tendency of knocking to occur is determined based on the magnitude of the knocking rate calculated by the knocking rate calculating unit. Engine combustion diagnostic device.   The regression line setting means obtains a regression line in a coordinate system having a power generation load on the horizontal axis and the in-cylinder pressure on the vertical axis, and the combustion state determination means is below a threshold line set below the regression line. The generator according to claim 5, further comprising: a misfire rate calculating unit that calculates a misfire rate based on the occurrence rate, wherein the tendency of misfire occurrence is determined based on the magnitude of the misfire rate calculated by the misfire rate calculating unit. Engine combustion diagnostic device.   The in-cylinder pressure detection value and power generation load data of the power generation engine are accumulated, and the load in-cylinder maximum pressure data representing the relationship between the power generation load and the in-cylinder maximum pressure at a predetermined time interval is accumulated, and an analysis data file is created. A remote terminal computer to be created; a combustion diagnostic database server for storing the analysis data file transferred from the remote terminal computer; and the maximum in-cylinder pressure data for a plurality of cylinders for a predetermined number of days from the combustion diagnostic database server. Read out, obtain a regression line indicating the correlation between the power generation load and the in-cylinder maximum pressure based on the data, set a threshold straight line based on the standard deviation up and down around the regression line, Combustion diagnostic device for determining a sign of abnormal combustion based on the occurrence rate with the maximum in-cylinder pressure, and diagnostic result of the combustion diagnostic device A WEB server to be stored and a client computer that provides the diagnostic result from the WEB server, the remote terminal computer, the combustion diagnostic database server, the combustion diagnostic device, the WEB server, and the client computer Are connected via a network to a combustion diagnosis system for a generator engine.

Images