KR100882887B1 - A trend monitoring and diagnostic analysis method for a vacuum pump and a trend monitoring and diagnostic analysis system therefor and computer-readable storage media including a computer program which performs the mothod - Google Patents

Abstract

The present invention, such as an active diagnostic algorithm, has been developed for the realization of the initial detection of the degradation of the vacuum pump to protect the pump from failure as well as for the maintenance of the forecast of the vacuum pump. According to the present invention, a simple and efficient way to deal with technical problems resulting from large variability in pump-to-pump operating characteristics and multiple process conditions, in particular in the semiconductor fabrication process, is made possible.

Description

A computer-readable storage medium comprising a trend observation and diagnostic analysis technique of a vacuum pump, the observation and analysis technique system, and a computer program for performing the technique. AND DIAGNOSTIC ANALYSIS SYSTEM THEREFOR AND COMPUTER-READABLE STORAGE MEDIA INCLUDING A COMPUTER PROGRAM WHICH PERFORMS THE MOTHOD}

The present invention relates to a diagnostic technique for preventing and maintaining a vacuum pump, and more particularly, to a semiconductor fabrication process having multiple operating conditions.

The demand for functionality and stability for vacuum pumps in modern semiconductor production processes has steadily increased. This demand is increased because as the size of the production wafer increases, the cost due to defective wafers and lost production time increases further. Technical requirements for vacuum pumps for such modern semiconductor processes are well pointed out by Bahnen and Kuhn [Ref. 1: R. Bahnen and M Kuhn, "Increased reliability of dry pumps due to process related adaptation and pre-failure. warning, "Vacuum, Vol. 44, No 5-7, pp. 709-712,1993]: high stability, high water solubility capable of pumping corrosive and reactive gas mixtures, unacceptable downtime, high water solubility and low vibration, low noise, etc. which can pump particulate and sublimable gas mixtures. To meet these demands, new dry pumps for modern semiconductor processes must provide water solubility suitable for operating conditions that depend on various processes [Ref. 1]. Application to the individual process has been shown to require measurement methods and adjustments dedicated to operating parameters such as temperature and gas pressure inside the dry pump stage. Such process-related parameters are very important in determining whether the operating conditions of the manufactured pump are satisfactory. In addition to process-related parameters, parameters related to pump operation (power, coolant, purge gas, wear of pump parts-bearings, seals) to apply the process to the vacuum pump and to avoid unplanned downtime (seals, gearboxes and motors) are also proposed by Bahnen and Kuhn [Ref. 1]. A monitoring plan for each operating parameter based on warning and alarm levels has been proposed to avoid unexpected pump failures. However, no reasonable way of selecting all warning and alarm levels of process dependent and operationally relevant parameters has been proposed. Such threshold level selection is still a very challenging issue in detecting early failure of the vacuum pump.

Initial level-based monitoring has been widely known as a traditional technique for protecting pumps from failures [Reference 2: R.H. Greene and D.A. Casada, Detection of pump degradation, NUREG / CR-6089 / ORNL-6765, Oak Ridge National Laboratory, 1995]. However, Wegerich et al. [Reference 3: S.W. Wegerich, D. R. Bell and X. Xu, “Adaptive modeling of changed states in predictive condition monitoring,” WO 02/057856 A2, 2002; Reference 4: S.W. Wegerich, A. Wolosewicz and RM Pipke, “Diagnostic systems and methods for predictive condition monitoring,” WO 02/086726 A1, 2002, pointed out the shortcomings of an initial warning and alarm plan based on the output of the sensor: " Existing technologies have not responded to global changes in the operating parameters of a process or machine, and sometimes failed to provide adequate warnings to prevent unexpected shutdowns, equipment damage or significant safety risks. " To overcome such limitations in existing technologies, they use a parametric model based on neural networks suitable for new operating conditions [Ref. 3] and an analytical system based on the model for monitoring predictive conditions [Ref. .4] was proposed. As is known from previous studies on emotion and coordination of dynamic systems [Reference 5: Wan-Sup Cheung, “Identification, stabilization and control of nonlinear systems using the neural network-based parametric nonlinear modeling,” Ph.D. Thesis, University of Southampton, 1993], neural network models are useful for interpolating new states that lie between training data sets and for extrapolating adjacent states that are very close to the trained set. Have the ability. Wegerich et al. [Ref. 3, Ref. 4] utilize the learned neural network insertion and extrapolation capabilities [Ref. 5] to estimate the current state of a process or machine corresponding to the sensor's output measurements. Estimated state values are extracted from the actually measured sensor output to obtain an error signal that is used to determine how far the process or system is from the model state. Moreover, these differences also generate initial threshold alerts, perform statistical tests to determine whether the process or system has changed to new operating conditions, and design new learning sets for changed operating areas. Is used. In addition, the error between the estimated state value and the measured state value can cause an early warning of an error, perform statistical tests to investigate changes when the process or system is operating under new conditions, or for new operating areas. It is used for the reconstruction of the learning set. The proposed signal processing plan, which includes the design of a new learning set for a modified operating region and a learning set for a model learning process of the learning set, and which generates an alarm and is suitable for the modified operating region, requires difficult computational work, Even the inherent complexity of the proposed model-based analysis system appears to be involved. The complexity of this unrealistic computational output and execution of the proposed inspection system is inevitable and encountered in pump observation systems for modern semiconductor fabrication processes. It became a technical issue. As a result, a simple model suitable for pump operating conditions is of great importance in developing a pump observation system. This aspect is one of the main technical issues of the present invention to be described later.

Instead of using the above parameter model suitable for changing the operating conditions of the vacuum pump over time, Ushiku et al . [Ref.6: Y.Ushiku, T.Arikado, S.Samata, T.Nako, and Y.Mikata, "Apparatus for predicting life of rotary machine, equipment using the same, method for predicting life and determining repair timing of the same," US Patent Application Publication, US2003 / 0009311 A1,2003], Samata et al. And Ref. 7: S. Samata, Y. Ushiku, K. Ishii, and T. Nakao, "Method for diagnosing life of manufacturing equipment using rotary machine," US Patent Application Pulblication, US2003 / 01543997 A1,2003] Ishii et al. [Ref.9: K.Ishii , T. Nakao, Y. Ushiku, and S. Samata, "Method for avoding irregular shutoff of production equipment and system for irregular shutoff," US Patent Application Publication, US2003 / 0158705 A1,2003. Determine if there is a difference from a standard data set of time series values corresponding to a condition A statistical analysis method and an analysis method based on mahalanobis distance have been proposed [Ref.10: WH Woodall, R, Koudelik, ZGStoumbos, KL Tsui, SB Kim, CP.Carvounis, "A review and analysis of the Mahalanobis-Taguchi system, "TECHNOMETRICS, Vol. 45, No. 1, pp 1-14,2003]. The statistical analysis methods are based on secondary statistical characteristics of sampled signals such as mean value, standard deviation, correlation, etc. (Ref.11: JSBendat AGPiersol, Random data: Analysis and measurement procedures, John Wiley & Sons: NY, 1985). . Since statistical properties can only be used in normal processes, such a scheme has limited applicability to the various output dependent operating conditions required for other products. This means that for each output dependent task a standard data set of each time series value is required. A major issue here is how to design a standard data set of output dependent time series values sufficient to cover all ranges of normal operating conditions. No efficient way of designing such data sets has been proposed by Ushiku et al. [Ref. 6], Samata et al. [Ref. 7, Ref. 8] and Ishii et al. [Ref. 9]. To overcome the limited ability of the technology to detect abnormal operating conditions using statistical analytical tools, they are also referred to as multivariate statistics [Ref.11: WHWoodall, R. Kooudelik, ZGStoumbos, KLTsui. SBKim, CP Carvounis, "A review and analysis of the Mahalanobis-Taguchi system," TECHNOMETRICS, Vol. 45, No. 1, pp. The mahananobis distance analysis method, which is well known in [1-14,2003], was considered for the quantitative analysis of the similarity between current time series data values and standard data values. When the standard time series data values cover the full range of normal operating conditions, the values obtained appear to be more efficient than the secondary statistical (average and deviation) methods used in existing trend observation systems. However, time-series data at normal operating conditions for new or reconditioned vacuum pumps are only available at the very beginning of each planned process, making it a standard for the full range of normal operating conditions without time-consuming data collection and signal processing. Data could not be obtained. Even in the modern semiconductor fabrication community, no method of designing such standard data has been well recognized. Indeed, the manufacturing units of modern semiconductors require a variety of processes at individual operating conditions such as various camber pressures, gas flow rates and gas mixtures, and gas characteristic values. The process-related characteristics and operating conditions in these semiconductor fabrications are incredibly confidential among vacuum pump suppliers. It is very important to note that vacuum pump observation and analysis systems for modern semiconductor processes must be self-adapting to various operating conditions. It is essential for dry pump observation and diagnostic systems for modern semiconductor processes to break down individual process conditions and develop active measures to diagnose the working conditions of those conditions. The present invention has been shown to provide a realistic solution to these technical issues later.

The inventors of this technique have already developed an accurate performance test and evaluation method for low vacuum pumps and have published their experimental results in several technical papers [Ref. 12: JYLim, SHJung, WSCheung, KHChung, YHShin]. , SSHong, and WGSim, "Expanded characteristics evaluation for low vacuum dry pumps," AVS 49th International Symposium, xx, 2002; Reference 13: JYLim, WSCheung, JHJoo, YOKim, WGSim, and KHChung, "Characteristics evaluation practice of predictable performance monitoring for low vacuum dry pumps," AVS 50th International Symposium, 9-10, 2003; Reference 14: W. S. Cheung, J. Y. Lim and K. H. Chung, "Experimental study on noise characteristics of dry pumps," Inter-noise 2002, Port Lauddale: USA, 2002; Reference 15: W. S. Cheung, "Acoustic characteristics of dry pumps designed for semiconductor processes," Inter-noise 2003, Jeju, Korea, 2003]. Such experiments were performed by a low vacuum pump test bench, the overview of which is shown in FIG.

The test bench is used to determine the performance factors of low vacuum pumps such as pumping speed (volume flow rate), ultimate internal pressure, compression ratio, gas output simulation, maximum and minimum values of operating pressure, power consumption, residual gas analysis and mechanical noise, vibration levels, etc. Has been. Hundreds of vacuum pumps supplied to semiconductor manufacturers have been tested. The test results of these vacuum pumps provided a systematic understanding of the key performance factors and dynamic characteristics of the vacuum pump.

2 shows statistical characteristics (maximum value, minimum value, average value) of the pumping speed measured from a plurality of pumps of the same model. Square, asterisk, and circular notation lines represent the maximum, minimum, and average values of the test results of the pumping speed, respectively, which test individual gas simulated by adjusting the internal gas pressure of the test dome shown in FIG. Obtained from output conditions. The coefficient of variability of the pumping speed, defined as the ratio of the standard deviation to the mean value, is found to be 6.7% when the internal pressure of the test dome is equivalent to 0.01 mbar, and 6.7% corresponding to an internal pressure of 0.002 mbar. It became.

At levels above 0.05 mbar, the coefficient of variation was found to be 3.5% or less. This means that within a small variation, the quality control of the pumping speed of the tested vacuum pump is very good. It is noteworthy that the pumping speed with small variability is a good indicator of whether the pump is currently operating. Indeed, the pumping speed is the most important factor among the performance parameters of low vacuum pumps. However, previous inventions for observing the operating conditions of a vacuum pump do not consider the pumping speed as the observed state variable.

In the next section, the present invention will disclose a systemic approach for observing the pumping speed of a pump installed in a semiconductor fabrication plant. 3 shows how much mechanical noise and vibration level variation is present between the tested pumps. Although the pumping speeds of the tested pumps had small variability as shown in FIG. 2, the mechanical noise and vibration levels were very different for each pump of the same model. The mechanical noise level is estimated by taking the average of the sound pressures measured from the ten positions recommended by the ISO 3744 standard. The maximum difference in the mechanical noise level values was observed to be 12 dBA when the test dome pressure was 2 mbar. Under other gas output conditions, the difference in sound pressure level value SPL approached about 9 dBA. This large SPL difference corresponds to a 4 times loudness difference (2 times the loudness per SPL difference of 5 dBA). The coefficient of variation of the sound pressure level was found to be 51% to 65% when the pressure range was 0.01 mbar to 10 mbar or more. The coefficient of variation of the mechanical vibration (acceleration) level was found to be 19% to 23% when the pressure was below 1 mbar and increased to 51% as the gas pressure reached 10 mbar. Moreover, the ratio of the minimum level value and the maximum level value of the mechanical acceleration level was 1.4 to 1.6 when the gas pressure was 1 mbar or less, but it was observed that the ratio increased rapidly to 3.3 as the gas pressure reached 10 mbar. This great change points out that each individual pump has its own normal operating conditions for mechanical noise and vibration. This pump-by-pump working characteristic can lead to unstable and inconsistent state observations, such as false alarms and alarm signals, even for a normally functioning instrument. The use of a fixed level based initial level threshold for generating a warning or alarm signal for the system has caused great difficulty. In order to improve such limited capabilities of fixed level based machine condition observation and diagnostic systems, the present invention will propose in section 3 an active algorithm that is self-adaptable to normal operating conditions that depend on pumps.

It should be noted that such large variability in the state variables of the machine operated observation system is not limited to noise and mechanical vibration signals. 4 shows the statistical characteristic values (maximum value, minimum value, average value) of the power consumption values measured from the booster pump and the dry pump. The maximum and minimum ratio of power consumption for the booster pump was 1.3 at a gas pressure of 2 mbar or lower, and increased to 1.6 as the gas pressure reached 10mba. The coefficient of variation of the booster pump was 9% to 11% when the gas pressure was below 1 mbar, but rose steeply to 57% as the gas pressure reached 10 mbar. In contrast to this large deformation of the power consumption of the booster pump, the maximum and minimum ratio of the power consumption value of the dry pump was observed to be 1.1 to 1.2 within the pressure range of the test. The coefficient of variation was also found to be 4% to 6% within the test pressure range. These test results indicate that the total power consumption value is useless for the state observation system because the total power consumption value of the booster pump and the dry pump is a highly variable state variable. As a result, two separate power consumption values of the booster pump and the dry pump will be considered in the present invention.

It is very important to understand how much the measured state variable values increase as the gas output conditions vary in the pump operating area. The experimental results presented in FIGS. 3 and 4 help to find a solution to this task by carefully observing the mean values (indicated by the solid lines marked with asterisks). Although the gas pressure in the test dome gradually increases to some extent, the average value remains the same. This is the area where the measured state values of mechanical noise and vibration and power consumption levels are independent of the gas output. The present invention does not utilize characteristics independent of gas output among the state variables measured for analyzing the operating conditions of the vacuum pump. Such gas output independent conditions are very often found in actual process conditions. A good example is the "idle" state of a working vacuum pump, in which no external gas is supplied to the inlet of the pump. In the next section, the present invention will propose a systemic method for modeling the output independent characteristics of the state variables of the vacuum pump observation and diagnostic system.

Moreover, as the gas pressure increases in a region independent of gas output, the mean values of mechanical noise, vibration and power consumption parameters increase. For example, the maximum value of the mechanical noise level in the gas output dependent region appears to be 12 dBA (4 times) higher than the maximum mechanical noise level in the gas output independent region. Similarly, the maximum value of the mechanical vibration level is 2.4 times higher in the region depending on the gas output, and the power consumption level values of the booster pump and the dry pump are 2.3 times and 1.2 times higher, respectively. Here, another technical issue encountered in state observation and diagnostic systems is finding a suitable model to account for such gas output dependent characteristics of state variables, which means that the actual working area of the vacuum pump always includes a gas output dependent condition. Because. In the next section, the present invention will also propose a systemic way of modeling the dynamic properties of state variables in the region depending on gas output. Of course, mathematically identical models have been shown to be applicable to both gas independent and dependent conditions. As a result, one model will be provided for the operating area independent of the gas output and the other model will be provided for the area dependent on the gas output. The use of two separate models has been developed to increase the stability and reliability of detecting abnormal operating conditions as soon as possible in a vacuum pump.

In the present invention, the observed information regarding the gas output condition, such as the inlet gas pressure signal of the vacuum pump, is characterized by distinguishing abnormal operating conditions of the vacuum pump, and more specifically, the increase in the observed state variables due to the output of the gas. It is very clear that it plays an important role in judging whether or not it is. In order to improve the ability to more reliably diagnose abnormal operating conditions of the vacuum pump, using the pressure information of the intake gas has not been devised in the previous inventions. In the present invention, observing the pressure of the intake gas seems to enable the quantitative analysis of the pumping speed with the improvement of the diagnostic ability. The above point is very important because finding an indication of the pumping speed leads to a suitable time to replace the target vacuum pump with a new one. The present invention provides a reasonable way of evaluating the pumping speed of a vacuum pump operating in a semiconductor fabrication plant.

Technical solution

According to the present invention, there is provided a trend observation and diagnostic analysis method for failure prevention and forecast maintenance of a vacuum pump, which includes the following steps and operates under an alternating state of idle and gas output conditions.

Sampling the time series data of the state variable signal at a predetermined rate under idle conditions and gas output operating conditions;

Classifying the maximum and minimum values of the time series data of the state variables from each subset of successively sampled signals over a period longer than a period of the varying state variable signal element under idle conditions and gas output operating conditions;

Using an active diagnostic algorithm based on a linear parameter model under idle conditions and gas output operating conditions, an optimized set of model parameters of the upper asymptotic boundary is estimated from the classified maximum values of each state variable, and the Estimating another optimized set of model parameters of the lower asymptotic boundary from the classified minimum values;

Each time a transition state from a gas output operating state to an idle condition is observed, using a in-situ evaluation method, obtaining a pumping speed index based on suction pressure;

Store the upper model parameter and the lower model parameter of the upper asymptotic and lower asymptotic boundary of all target state variables in each idle condition and gas output operating condition, and each time a transition state from the gas output operating state to the idle condition is observed, Storing the obtained pumping speed indicator;

In idle conditions and gas output operating conditions, repeating estimating model parameters of upper and lower boundaries of each state variable and obtaining pumping indicators;

Observe the drift trend from the estimated upper and lower model parameters of all target state variables collected under continuous idle operating conditions, and view the drift trend from the estimated upper and lower model parameters of all target state variables collected under continuous gas output operating conditions. Observing a fluctuation trend from a pumping speed indicator continuously collected from the transition from the gas output operating state to the idle state, and diagnosing whether or not the vacuum pump is abnormal based on the fluctuation trend analysis result;

According to the present invention, there is also a trend observation and diagnostic analysis system for failure prevention and forecast maintenance of a vacuum pump, which is composed of the following components and operates under multiple processes including an alternating state of idle and gas output conditions. Is provided.

A data acquisition unit for sampling in real time the time series data of the state variable signal at a predetermined speed in the idle phase and the gas output phase;

Means are provided for calculating and storing the measured signals from said data collection unit, characterized by providing the following steps;

Classifying the maximum and minimum values of the time series data of the state variables from each subset of successively sampled signals over a period longer than a period of the varying state variable signal element under idle conditions and gas output operating conditions;

Using an active diagnostic algorithm based on a linear parameter model under idle conditions and gas output operating conditions, an optimized set of model parameters of the upper asymptotic boundary is estimated from the classified maximum values of each state variable, and the Estimating another optimized set of model parameters of the lower asymptotic boundary from the classified minimum values;

Each time a transition state from a gas output operating state to an idle condition is observed, using a in-situ evaluation method, obtaining a pumping speed index based on suction pressure;

Storing the upper model parameter and the lower model parameter of the upper asymptotic and lower asymptotic boundary of all target state variables in each idle condition and gas output operating condition, and storing the obtained pumping speed index;

Observe the drift trend from the estimated upper and lower model parameters of all target state variables collected under continuous idle operating conditions, and view the drift trend from the estimated upper and lower model parameters of all target state variables collected under continuous gas output operating conditions. Observing a fluctuation trend from a pumping speed indicator continuously collected from the transition from the gas output operating state to the idle state, and diagnosing whether or not the vacuum pump is abnormal based on the fluctuation trend analysis result;

Favorable effect

The most challenging issue in the present invention has been to find a simple and efficient way to deal with the technical issues arising from the large variability of pump by pump operating characteristics and multiple process conditions.

The present invention proposes two separate model parameter sets, one estimated under the idle operating conditions of the vacuum pump and one estimated under the gas output operating conditions. This means that the measured signals of the state variables under two operating conditions have quite different statistical characteristics, and the diagnostic analysis for idle and gas output operating conditions is adopted to realize a greatly improved performance for early detection of failure of the vacuum pump. Because it becomes.

Of course, the present invention provides an efficient way to distinguish between idle and gas output operating conditions using measured state variables such as suction pressure, booster pump supply current and exhaust pressure.

Furthermore, the present invention introduces an in-situ algorithm for calculating the pumping speed by using the measured suction pressure, and the proposed algorithm for calculating the pumping speed investigates how much the current pumping speed is lower than the initial value. Makes it possible to do This knowledge of the lowering level of the pumping speed is very valuable as it allows the maintenance engineer to decide when to replace the target vacuum pump with a new one.

Furthermore, the present invention provides a reasonable way to design a series of optimized model parameters and pumping speed indicators corresponding to seven state variables into matrix type data suitable for multivariate statistical analysis, performance analysis and Mahana Novis distance analysis. . The conversion of the data matrix of the model parameter structure to the above existing analysis algorithms (multivariate statistical analysis, performance analysis and Mahana nobis distance analysis) is also one of the main achievements contributed by the present invention.

1 shows a schematic diagram of a performance test bench of a low vacuum pump according to the present invention.

Figure 2 shows the statistical characteristics for the pumping speed of the low vacuum pump.

3A and 3B show the noise level characteristics and the mechanical vibration level characteristics of the spatially averaged low vacuum pump, respectively.

4A and 4B show power consumption characteristics of the booster pump and the dry pump, respectively.

5A-5D show measured state variable signals, suction pressure and exhaust pressure, and supply current, respectively, for boosters and dry pumps.

6A-6D show a comparison of the classified maximum, minimum (thin solid line) of the inlet and exhaust pressures of the booster and dry pump and current signals and the estimated results (solid solid line) based on a suitable model, respectively.

7A to 7D show the vibration mean acceleration and noise mean square values (rms) and their upper and lower asymptotic boundary curves (bold solid lines), respectively.

FIG. 8 shows the suction pressure signal of the first negative going transition region existing between the first gas output operating region and the second idle region. (Bold solid lines optimize the exponentially decaying function.) FIG. Point to the model.)

Active Diagnosis Algorithm of Vacuum Pump

The state variable of the present invention is defined as one of the periodically sampled physical parameters selected for quantitatively examining the operating conditions of the subject vacuum pump. State variables related to pump operation include various variables such as motor current, suction pressure and exhaust pressure, negative pressure signal, mechanical vibration signal, purge gas pressure and gas flow rate, body temperature, coolant temperature, lubricant pressure and level of vacuum pump. have. The above vacuum pump state variables have been used to diagnose the operating conditions of the vacuum pump. However, only a limited number of state variables have been selected for efficient diagnosis of vacuum pumps. The choice of state variables in the present invention was made according to the gas output dependency. If the response of the state variable to the suction pressure of the vacuum pump is highly dependent, it is classified as a "real time" observation variable sampled over each short period (ie 10 samples per second). As real-time observation variables, suction and exhaust pressures, supply currents of booster pumps and dry pumps, mechanical vibration and noise signals are considered in the present invention. On the other hand, if the response dependence of the variable is significantly low, it is classified as an "auxiliary" state variable sampled at a low rate (ie 1 sample per second). As auxiliary state variables, purge gas pressure and gas flow rate, body temperature, coolant temperature, lubricating oil pressure and level of vacuum pump are handled in the present invention. Since the existing two-dimensional statistical scheme is sufficient to successfully implement the observation and diagnosis of the auxiliary state variable, the present invention does not consider this auxiliary state variable. This is because the auxiliary state variables satisfy statistical static conditions well. Future inventions may investigate the effectiveness of the proposed active diagnostic algorithm for state observation and diagnostic analysis for the auxiliary state variables listed above.

Digital Signal Processing and Conditioning Theory [Ref. 16: B. Windrow and S. D. Steams, Adaptive Signal Processing, Prantice-Hall, Englewood Cliffs: NJ, 1985; Ref.17: P.A.Nelson and S.J. Elliott, Active Control of Sound, Academic Press, London, England, 1992], an active algorithm, provides an efficient means to adjust the parameters of a selected target system model to account for the mechanical properties inherent in measured state variables. do. The active algorithm allows the estimation of model parameters that are well adapted to dynamically varying state variables. The estimated model parameters are utilized to diagnose the operating conditions of the vacuum pump. This theoretical approach is referred to herein as an "active diagnostic" algorithm. It is important to note that the active diagnostic algorithm provides a set of model parameters tailored to various pump operating characteristics, ie multiple process conditions. Of course, the active algorithm also allows estimation of model parameters for other pumps. This set of pump dependent parameters is very useful for investigating the operating variability of groups of the same model vacuum pump. This is the reason for using an active algorithm based on a parametric model for the diagnosis of vacuum pumps.

1. Active Algorithm for Parameter Modeling of State Variable Diagnosis

Selection of the parametric model in the present invention was made by observing the signal characteristics of the measured state variables. 5 shows measured signals of the suction pressure 5 (a), the exhaust pressure 5 (b), the supply current 5 (c) of the booster pump, and the supply current 5 (d) of the dry pump, respectively. As shown, they were sampled at a rate of 10 words per second.

As shown in Fig. 5A, two separate amplitude regions, a group of variable amplitude level regions and a group of uniform amplitude level regions, are clearly observed from the suction pressure signal. The equal amplitude level interval corresponds to the idle operating state of the vacuum pump in which no pumping gas is supplied externally from the process chamber.

The fluctuating amplitude level region corresponds to the pumping operation state under the gas output condition which varies between the upper asymptote and the lower asymptote shown in Fig. 5A. Let ym denote the m-th sampled suction pressure signal and the subscript m denote the time index. In the present invention, the sampling rate was selected to 10 Hz (10 samples per second). Sampled time series data {ym: m = 1,2, ...} of the suction pressure are used to classify the minimum and maximum values at a user-selected period, ie every 30 seconds or every minute. Is chosen to be longer than the period of the suction pressure signal that fluctuates under gas output operating conditions. Since the longest period of the fluctuating pressure signal is 54 seconds, the classification of each maximum and minimum value was done every minute.

6 shows the classified maximum and minimum values (thin solid line), the suction pressure signal 6 (a), the exhaust pressure signal 6 (b), the supply current signal 6 (c) of the booster pump, and the dry pump. A comparison of the estimation results based on the model optimized for the supply current signal 6 (d) is shown.

6 (a) shows the classified maximum and minimum values of the suction pressure signal. The maximum and minimum values are set to {yU, n, yL, n: n = 1,2, ...} obtained from each set of 600 consecutively sampled signals (corresponding to 1 minute recording signals). . The present invention proposes the use of a linear model for the upper and lower asymptotes, the linear model is as follows.

In Equation 1, the subscript k represents an upper or lower asymptote model, i.e. k = U for the upper model and k = L for the lower model. In Equation 1, two sets of {a _k , b _k : k = U or L} model parameters are easily obtained by using the least squares method. In the first gas output state let the time series values of the classified maximum and minimum values be {y _{k , n} = 1,2, ... N}. The optimized model parameters are found as follows.

The first parameter {a _k : k = U or L} in Equation 2 is the slope of the suction pressure signal indicating the increase or decrease rate. The second parameter {b _k : k = U or L} indicates each initial suction pressure level (ie the value at n = 0). The thick solid line in FIG. 6 shows the values obtained from a suitable model for the upper and lower asymptotes. The estimated upper and lower asymptotes appear to be appropriately formed at the upper and lower boundaries of the measured amplitude signal. Moreover, the estimated model parameters can be used to investigate how much change in suction pressure is present at the first gas output operating condition. This indicates that the trend of suction pressure can be quantitatively characterized by the estimated model parameters. The above points are of great value, as two sets of appropriate model parameters enable trend observation and diagnosis for the measured suction pressure conditions. Since the proposed approach does not use all the set of sampled time series data, the use of appropriate model parameters provided a lot of memory savings. This means that implementation systems based on small hardware can be realized by using the appropriate model parameters. It should be noted that the mean and standard deviation at each of the upper and lower asymptotes can be obtained by using the equation

If the slope value is zero (a = 0 in Equation 3), the second parameter appears to be an average value.

Equation 3 shows that the estimated parameters enable calculation of statistical characteristics (average and standard deviation values) for the estimated model and analysis of the rate of increase or decrease of the measured suction pressure. The usefulness of using a parametric model for the statistical properties of This statistical information on the suction pressure signal is very useful for investigating the current gas output conditions of a working vacuum pump. Active diagnostics were modeled to model the mechanical operating characteristics observed from gas output operating conditions. As shown in Fig. 5 (a), the magnitude of the suction pressure appears to be uniform, but there appears to be some variation when the magnitude of the even amplitude level zone is reduced. Two parametric models corresponding to the upper and lower asymptotic boundaries were also considered for the idle state. The same active algorithm that was implemented for gas output pumping conditions was also used to estimate the model parameters of the upper and lower asymptotes for the idle state. Given the time series values of the classified maximum and minimum values of the sampled suction pressure signal under idle conditions, two sets of upper and lower model parameters are obtained using equation (2). In addition to the estimated model parameters and statistical characteristic values of the parameters under gas output operating conditions, the appropriate model parameters and their characteristic values, estimated in the idle operating region, are also used for trend observation and diagnosis of the vacuum pump. The combined set of parameters in the idle and gas output operating conditions not only examines how many gas output conditions have been applied on the vacuum pump, but also helps determine how much vacuum level is maintained in the idle state. useful. Knowledge of the gas output conditions of a vacuum pump is used to identify the cause of the observed abnormal operating conditions, i.e. to determine whether the abnormal pump operation is caused by abnormal gas output conditions or by other mechanical faults. It will seem to play a very important role. If such abnormal pump operation is caused by any unusual gas output condition, this is not the case with the vacuum pump being defective. Therefore, knowledge of the gas output conditions of a working vacuum pump is very important for accurate and stable diagnosis. The present invention highlights in detail the use of suction pressure signals for the precise diagnosis of vacuum pumps used in semiconductor fabrication processes.

The theoretical background of the parametric model, adopted to explain the dynamic characteristics of the suction pressure signal observed from the vacuum pump, was mentioned above. This approach is also applicable to other state variables such as the exhaust pressure signal shown in FIGS. 5 (b) to 5 (d) and the supply current signals of the booster and dry pumps. The time series values of the maximum and minimum values for each state variable were easily obtained by classifying the maximum and minimum values from every block of 600 consecutive samples (corresponding to the signals recorded per minute) supplied from the data acquisition system. . 6 (b) to 6 (d) show estimation results based on a model suitable for the classified maximum and minimum values (thin solid line) and exhaust pressure signals and supply current signals of the booster pump and the dry pump (bold solid line). do. Given the classified time series values of each state variable for the idle and gas output operating conditions, two sets of parameters corresponding to the upper and lower boundaries are obtained by using Equation (2). An estimated set of parameters at the upper and lower boundaries can also be used to investigate how much variation of each state variable is maintained under repeated idle conditions and gas output operating conditions.

5 and 6, it is shown that the pump operating conditions observed for 5 hours consisted of four operating stages, namely four idle states and three gas output states. Table 1 shows an estimated set of parameters for the upper and lower asymptotic boundaries at each operating condition: four idle states (idle states 1 to 4 in Table 1) and three gas output states (gas output states 1 to 3 in Table 1). ).

Table 1 shows the estimated parameter set of the upper and lower asymptotic boundaries of the four state variables (suction pressure, exhaust pressure, current signals supplied to the booster pump and the dry pump). BP and DP mean booster pump and dry pump, where a _U and b _u are The slope and initial value of the upper boundary asymptote, a _L And b _L means the slope and initial value of the lower boundary asymptote.

Note that the transition state between the idle state and the gas output state is not used for parameter estimation. The time intervals used for parameter estimation are detailed in the third row of Table 1. It is clear from Table 1 that the properties of each state variable are described by six parameters, each of which has two time stamps (initial and last time), two model parameters for idle (tilt) And two model parameters (tilt and initial value) for the gas output state. The parameters indicate how the suction pressure varies between the upper and lower boundaries under these two states when an idle state gas output condition occurs. Of course, comparison of successive idle states (ie odd numbered steps: steps 1, 3, 5, 7 of Table 1) appears to enable quantitative analysis of their variation. Moreover. Quantitative trend analysis between successive gas output states (ie, even-numbered stages: stages 2, 4, and 6) is shown by comparing the modal parameters of each state. Such a trend analysis of the other state variables, ie the exhaust pressure and the supply current of the booster pump and the dry pump, is easily implemented by comparing the mode parameter corresponding to each of the state variables listed in Table 1. The proposed diagnostic method provides an active algorithm for the estimation of model parameters suitable for the operating conditions of the vacuum pump, and it is obvious to use the model parameters for the trend analysis of each state variable.

It is shown in the present invention that the operating characteristic values are obtained for two separate idle and gas output operating regions. As shown in Figs. 5 and 6, the four state variables, suction pressure, exhaust pressure and the upper and lower boundary levels of the supply current of the booster pump and the dry pump, are clearly dependent on the gas output conditions. Such mechanical and electrical state variables are generally considered to be a class of static characteristics. Unlike such static state characteristics, mechanical vibration and noise signals containing high frequency components have been used as state variables for trend and diagnostic analysis.

7 (a) to 7 (c) show the vibration acceleration of the booster pump and the mean square value (rms) of the noise signal measured near the intermediate position between the booster pump and the dry pump. The frequency bandwidth of the noise signal was selected from 10 Hz to 10 kHz, and the frequency bandwidth of the noise signal was selected from 20 Hz to 20 kHz. Both signals were digitally sampled at a rate of 40.96 kHz (ie 40,960 samples per second). Each block of 4096 samples (which corresponds to an interval of 100 ms) was used to calculate the root mean square (rms) values shown in FIGS. 7 (a) and 7 (c). Each record of 600 rms values (corresponding to the interval value of 1 minute) is used to classify the maximum and minimum values (thin solid line) of vibration acceleration and noise level shown in FIGS. 7 (b) and 7 (d). It became.

The model parameters of the upper and lower asymptotes of the vibration acceleration and noise level are listed in Table 2. Table 2 shows the estimated parameter sets of the upper and lower asymptotes of vibration acceleration and noise level, a _U and b _U denote the slope and initial value of the upper boundary curve, and a _L and b _L denote the slope of the lower boundary curve. And initial value.

As given in Fig. 1, the above model parameters were estimated from seven operating conditions (four idle operating states and three gas output states). The vibration acceleration and noise level of FIG. 7 are not shown to show notable gas output dependent characteristics, unlike the suction pressure, exhaust pressure and supply currents of the booster pump and the vacuum pump shown in FIGS. 5 and 6. The lower asymptotic boundary of the vibration acceleration and noise level is almost independent of the gas output condition, but their upper asymptotic boundary appears to reveal sign-change (positive and negative sign) characteristics of the parameters related to the asymptotic slope. Gas output operating conditions appear to correspond to slopes with positive sign values, but the first three idle operating states, except for the last idle region, appear to correspond to slopes with negative sign values. One of the distinguishing features shown in FIG. 7, namely the variable signal element, was observed from the upper boundary level. The peak value of the variation element for the seven separate operating steps, with the circular symbol marked in FIGS. 7 (b) and 7 (d), was selected as another diagnostic variable. Their level and corresponding time stamps are also given in Table 2. The level and time stamps are very useful in determining how much excessive vibration and noise levels have occurred during each operating phase. As a result, this makes it possible to ascertain in which process a possible warning and alarm condition has occurred.

It is very interesting that the set of parameters of the upper and lower asymptotic boundaries of each state variable can drastically reduce the size of the data used for trend observation and diagnostic analysis. The current version of the electronic diagnostic guidelines [Ref. 18: Harvey Wohlwend, e-Diagnostics Guidebook, International SEMATECH, Version 1.5, October, 2002] has a minimum sampling rate of 10 Hz (10 samples per second) or higher for each state variable. High is recommended. In the present invention, the sampling rate was selected at 10 Hz according to the electronic diagnostic guidelines. As mentioned previously, a signal sampled for 5 hours was adopted in the present invention. The total number of samples for each state variable is estimated to correspond to 18,000. In contrast, the task characteristic values and their time stamps obtained for each static state variable were found to be only 42 data sets (7 sets of 4 model parameters and 7 sets of initial and last time stamps). . If a dynamic state variable is considered, 14 additional data (seven sets of peak values and corresponding time stamps) are added to the data. The rate of reduction of this diagnostic analysis data is extremely high. These are described in the TMS320C2000 series [Ref. 19: Data Manual for TMS320LF2407, TMS320LF2406, TMS320LF2402 Digital Signal Processors, Literature Number: SPRS094I, September 2003; Ref.20: Trend Observation and Diagnosis System by using the smallest size digital signal processor like TI Model of Data Manual for TMS320F2810, TMS320F2811, TMS320F2812, TMS320F2810, TMS320C2811, TMS320C2811, TMS320C2812, Literature Number: SPRS174J, December 2003]. Enable implementation of

2. Pumping  Field Estimation of Speed

In the previous subsection, active algorithms for estimating model parameters suitable for operating conditions of vacuum pumps are mentioned in detail. It was shown that the model parameters of the upper and lower asymptotic boundaries were estimated under separate idle and gas output operating conditions. Reasonable measures for separating pump operating conditions are introduced in this subclause. Since the suction pressure in the present invention is measured directly, it is natural to be used for such separation of conditions. When the semiconductor fabrication process is maintained in the reaction chamber, the suction pressure level of the vacuum pump is maintained above the minimum level due to the gas flow rate supplied from the reaction chamber. The minimum level, referred to herein as the initial level, has been found to be dependent on the process gas and associated process. For example, the initial level of the suction pressure signal observed from FIG. 5 (a) exceeds 9 [mbar]. The present invention adopted an initial level of 5 [mbar] to safely separate idle and gas output operating conditions. This safe choice made no faulty judgment. In the present invention, the transition region is defined as the 10 second time interval until 10 seconds have elapsed from the initial level. The suction pressure signals sampled in the transition region will play a very important role in estimating the parameters related to the pumping speed later. Although the suction pressure signal is not measured immediately, the division of the pump operating conditions is not limited. The use of either the supply current signal or the exhaust pressure signal shown in FIG. 5 enables the selection of the initial level in the same way as was done to obtain the suction pressure signal. As shown in Figs. 5 (b) to 5 (d), since the supply current signal of the booster pump has a closer similarity to the suction pressure than the exhaust pressure signal or the supply current signal of the vacuum pump, the present invention provides a pump operating condition. We propose the use of the supply current signal of the booster pump as a second choice for the classification of. However, the use of exhaust pressure seems to require an electrically tuned amplifier and noise filter circuit to minimize the separation of defects as much as possible. The proposed method for distinguishing idle and gas output operating conditions is one of the differential achievements contributed by the present invention.

Two kinds of suction pressure transition zones, positive-going zones or negative-going zones, were observed in FIG. 5 (a). A positive-going transition of suction pressure occurs when the exhaust valve of the reaction chamber is opened by starting the performance test of the newly installed vacuum pump, and proceeding when the exhaust valve is almost closed after the gas output test phase is completed. A negative-going transition occurs. Since the above suction pressure signal characteristics are smooth, the present invention utilizes the suction pressure measured in the secondary advancing transition region similar to the pump down test method.

FIG. 8 shows the suction pressure signal of the first sub-transition region existing between the first gas output operating region and the second idle region, in which the thick solid line represents a suitable model of exponentially decaying function. Point.

In the disclosure of the present invention, it has been clarified that the localization characteristic of the suction pressure signal shown in FIG. 8 is directly related to the pumping speed of the installed vacuum pump. The basic relationship between pumping speed and pump-down times, well known in vacuum theory (Ref. 20: Nigel. S. Harris, Modern Vacuum Practice, McGraw-Hill Book Company, Lendon: England, 1989), was utilized in the present invention. The algebraic equation is

In Equation 4, the symbols Q and V mean pumping speed [m ³ / h] and volume [m ³ ], respectively. The symbol ΔT means the second sampling period (ΔT = 100 [ms] in the present invention). The symbol alpha in Equation 4 is an exponential decaying constant whose value is directly related to the pumping speed. The formula is the initial value P ₀ and the last level P _n of the pressure range. It is assumed that the pumping speed is constant. As a result, by adopting the linear region on the semi-log plots shown in Fig. 8, a range suitable for the initial and final suction pressure levels is determined. The thick solid line indicates the selected area for the initial position and the final position used to estimate the exponential decay constant in the adopted pressure region. In the first zone, the initial and final pressure levels were chosen to be 80% and 20% of the observed suction pressure, respectively, before the continuous gas flow stopped. In the second zone the initial and final pressure levels were chosen to be 9% and 7%, respectively. These guidelines for selecting initial and final pressure positions have proven to be very stable and efficient in estimating exponential decay constants from many field tests. As the semiconductor fabrication process proceeds, the estimated exponential decay constant is used to investigate how the pumping speed performance has decreased.

It is very straightforward to estimate the optimized exponential decay constant corresponding to the selected area. Let {Pn: n = 1, ..., N} be the suction pressure signal sampled in the selected area. The logarithm of the suction pressure signal is obtained as follows.

The optimized parameters alpha and beta are estimated by using the least-squares method mentioned in the previous section, as given in equation (2). The exponential decay constants optimized for the two separate regions are used to approximate the pumping speed index, which is defined as the pumping speed per unit volume, as follows.

 Table 3 shows the estimated exponential decay constants and the estimated pumping rate indicators corresponding to the constants for the three consecutive negative transition regions of suction pressure shown in FIG. 5 (a), where the symbols alpha and Ip are Respective decrement constants and pumping speed indicators, respectively.

By using the measured suction pressure signal, the in-situ method proposed in the present invention to obtain the pumping speed index can be determined by the pump maintenance engineer to determine whether the target vacuum pump should be replaced by determining how much the pumping speed has been reduced. The field approach is very important as it provides the appropriate information to make the decision. The proposed field method of estimating pumping speed indicators is a very unique method that has not yet been introduced in recent pump diagnostic techniques [Ref.1-Ref.4, Ref.6-Ref.9].

3. Using model parameters trend  Observation and Diagnosis Methods

The six state variables considered in the previous subclauses are the suction and exhaust pressures, the current supplied to the booster pump and the dry pump, and the mechanical vibration and noise signals. The active algorithm for estimating the model parameters suitable for the operating conditions of the vacuum pump is optimized model parameters {a _U , b _U , a _L , b _L } and mean values for the gas output operating conditions at each idle and gas output state. And peak values. As introduced in the second subsection, the peak value collected every minute, represented by {VU, PK}, is also added to the four parametric models for each (idle or gas output) operating condition. As a result, the five parameters {a _U , b _U , a _L , b _L , VU, PK} are representative data sets for each operating condition (idle state or gas output state). As the semiconductor fabrication process proceeds, a series of five parameters for all considered state variables will be depicted in the form of a two-dimensional matrix.

Note that the subscript symbols ("IDLE" and "LOAD") mean idle operating conditions and gas output operating conditions. The raw row index n means a sequence of manufacturing processes. The column indexes j and k denote a classification number of seven state variables and an order of five parameters for each state variable. Although the mechanical vibration signal measurement of the booster pump is not disclosed in the present invention, the seventh state variable corresponds to the vibration signal measurement. If necessary, a pumping speed indicator may be included in the last column of the matrix. The selection of the classification number and the parameter order can be done in any convenient way. When either the idle state or the gas output operating state is implemented, a corresponding horizontal column vector is obtained.

Statistical analysis of single or multiple variables, Mahana nobis distance analysis [Ref.10] and process performance analysis [Ref.21: Z.G. Stoumbos, "Proess capability indices: Review and extensions," Nonlinear Analysis: Real World Applications, Vol. 3, pp. By utilizing well-known analysis methods such as 191-210, 2002, the matrix data described in Equation 7 can be easily used for precise analysis of the target vacuum pump. Indeed, in this subclause, the concept behind the statistical analysis of a single variable is used to indicate how well the estimated characteristic values can be used for precision diagnostic analysis to maintain forecasts of low vacuum pumps. The concepts and logical sequence disclosed in the previous subclauses fit well with the statistical analysis of single variables. However, multivariate analysis, including process performance analysis and Mahana nobis distance analysis, has not yet been considered, since the technical discussion thereof is beyond the scope of the present invention. The present invention prefers Mahana nobis distance analysis over multivariate analysis or process performance analysis. Because the Mahana Nobis distance analysis gives us a more sensitive response to small changes in the estimated model parameters. Matrix data consisting of optimized model parameters and average peak values of dynamic characteristic values imparted to the measured state variables lead to another efficient means for precise analysis for predictive maintenance of vacuum pumps. The conversion of structured data matrices into existing analytical algorithms such as multivariate statistical analysis, process performance analysis, and Mahana nobis distance analysis is, of course, one of the contributions contributed by the present invention.

The present invention, such as an active diagnostic algorithm, has been developed for the realization of the initial detection of the degradation of the vacuum pump to protect the pump from failure as well as for the maintenance of the forecast of the vacuum pump.

According to the present invention, a simple and efficient way to deal with technical problems resulting from large variability in pump-to-pump operating characteristics and multiple process conditions, in particular in the semiconductor fabrication process, is made possible.

Claims (12)

Sampling the time series data of the state variable signal at a predetermined rate under idle conditions and gas output operating conditions; Classifying the maximum and minimum values of the time series data of the state variables from each subset of successively sampled signals over a period longer than a period of the varying state variable signal element under idle conditions and gas output operating conditions; One optimized model parameter of asymptotic upper bound from the classified maximum value of each state variable, using an active diagnostic algorithm based on a linear parametric model under idle conditions and gas output operating conditions. Estimating a set, and estimating another optimized set of model parameters of asymptotic lower bound from the classified minimum of each state variable; Each time a transition state from a gas output operating state to an idle condition is observed, using a in-situ evaluation method, obtaining a pumping speed index based on suction pressure; Store the upper model parameter and the lower model parameter of the upper asymptotic and lower asymptotic boundary of all target state variables in each idle condition and gas output operating condition, and each time a transition state from the gas output operating state to the idle condition is observed, Storing the obtained pumping speed indicator; In idle conditions and gas output operating conditions, repeating estimating model parameters of upper and lower boundaries of each state variable and obtaining pumping indicators; and Observe the drift trend from the estimated upper and lower model parameters of all target state variables collected under continuous idle operating conditions, and view the drift trend from the estimated upper and lower model parameters of all target state variables collected under continuous gas output operating conditions. Observing fluctuation trends from pumping speed indicators continuously collected from the transition from the gas output operating state to the idle state, and diagnosing whether or not the vacuum pump is abnormal based on the fluctuation trend analysis results, A trend observation and diagnostic analysis method for failure prevention and forecast maintenance of vacuum pumps operating under alternating idle and gas output conditions. The method of claim 1, The vacuum pump is a trend observation and diagnostic analysis method for the failure prevention and forecast maintenance of the vacuum pump, characterized in that used in the semiconductor manufacturing process. The method of claim 1, The state variables are the suction and exhaust pressures, the supply current of the booster pump and the dry pump, the level squared mean (rms) level of the mechanical vibration of the booster pump and the dry pump, and the noise mean square near the intermediate position between the booster pump and the dry pump. (rms) level trend monitoring and diagnostic analysis method for failure prevention and maintenance of the vacuum pump characterized in that it comprises a. The method of claim 3, In the case of the level squared mean (rms) level of the mechanical vibration of the booster pump and the dry pump and the noise mean square (rms) level near the intermediate position between the booster pump and the dry pump, the mechanical vibration for the idle condition and the gas output operating condition And estimating a peak value of the noise signal and the noise signal. The method of claim 1, The in-situ evaluation method is a standard value, in a manner similar to the pump down test method in which the negative going transition region of suction pressure is used to obtain the pumping speed. Trend observation and diagnostic analysis method for the failure prevention and forecast maintenance of the vacuum pump comprising a method for measuring the suction pressure signal. The method of claim 1, The pumping speed index is a trend observation and diagnostic analysis method for the failure prevention and forecast maintenance of the vacuum pump, characterized in that defined by the pumping speed per volume of Equation 1 below. (Equation 1)

In Equation 1, the symbols alpha and Ip mean exponentially decaying constant and pumping speed index, and the symbols Q and V mean pumping speed [m ³ / h] and empty volume [m ³ ], The exponential decay constant is obtained from the logarithm value of the suction pressure signal {Pn: n = 1, ..., N} corresponding to the negative going transition region defined in Equation 2 below. Saved. (Equation 2)

In Equation 2, the optimal exponential decrement constant alpha and initial value beta are obtained using the least square method. The method of claim 1, The linear parametric model for explaining the upper and lower asymptotic boundaries of each sampled state variable is given by Equation 3 below. (Equation 3)

In Equation 3, the subscript k means an upper or lower asymptotic model {k = U or L}, and the upper and lower model parameters {a _k , b _k : k = U or L} for each state variable. Both sets are obtained by using the least square method of Equation 4 below. (Equation 4)

In Equation 4, the first parameter {a _k : k = U or L} is the slope of the upper and lower asymptotes of each measured state variable, and the second parameter {b _k : k = U or L} is the upper and lower asymptotes. Each initial value of. The method of claim 1, The determination of the transition state from the gas output operation state to the idle condition is performed by observing the initial value of the suction pressure level, the trend observation and diagnostic analysis method for the failure prevention and forecast maintenance of the vacuum pump. The method of claim 1, The determination of the transition state from the gas output operation state to the idle condition is performed by observing an initial level of one of the supply current level and the exhaust pressure signal of the booster pump, for failure prevention and forecast maintenance of the vacuum pump. Trend observation and diagnostic analysis method. The method of claim 1, Observing and diagnosing the fluctuation trend of the target pump consists of a pair of two-dimensional structures of the following equation, consisting of estimated model parameters and peak values of each state variable obtained from each idle condition or gas output operating condition. Trend observation and diagnostic analysis method for failure prevention and forecast maintenance of a vacuum pump, characterized in that performed by using a data matrix.

for n = 1, ...; j = 1, ..., J (total number of state variables related to measured pump operation) k = 1, ..., K (total number of pump operating characteristic values) In the above equations, the subscripts "IDLE" and "LOAD" denote idle conditions and gas output operating conditions, the horizontal index n means a sequence of performance tests, and the vertical index j and k are states related to pump operation. The classification number of the variable and the order of the pump operating characteristic values for each state variable. A data acquisition unit for sampling in real time the time series data of the state variable signal at a predetermined speed in the idle phase and the gas output phase; Consisting of a signal processing unit characterized by providing the following steps providing a means for calculating and storing the measured signal from the data collection unit, The signal processing unit obtains the maximum and minimum values of the time series data of the state variables from each subset of successively sampled signals over a period longer than the period of the varying state variable signal element under idle conditions and gas output operating conditions. Sorting; Using an active diagnostic algorithm based on a linear parameter model under idle conditions and gas output operating conditions, an optimized set of model parameters of the upper asymptotic boundary is estimated from the classified maximum values of each state variable, and the Estimating another optimized set of model parameters of the lower asymptotic boundary from the classified minimum values; Each time a transition state from a gas output operating state to an idle condition is observed, using a in-situ evaluation method, obtaining a pumping speed index based on suction pressure; Storing the upper model parameter and the lower model parameter of the upper asymptotic and lower asymptotic boundary of all target state variables in each idle condition and gas output operating condition, and storing the obtained pumping speed index; Observe the drift trend from the estimated upper and lower model parameters of all target state variables collected under continuous idle operating conditions, and view the drift trend from the estimated upper and lower model parameters of all target state variables collected under continuous gas output operating conditions. And observing fluctuation trends from pumping speed indicators continuously collected from the transition from the gas output operating state to the idle state and diagnosing whether or not the vacuum pump is abnormal based on the fluctuation trend analysis result. A trend observation and diagnostic analysis system for failure prevention and forecasting of vacuum pumps that are characterized and operated under multiple processes including alternating idle and gas output conditions. A computer program for performing the trend observation and diagnostic analysis method according to any one of claims 1 to 10 for preventing failure and maintaining a forecast of a vacuum pump operating under an alternating idle condition and a gas output condition. And a computer readable storage medium.

Images