FIELD

The present technology is directed to a system for measuring and monitoring industrial machining equipment. More specifically, it is an autonomous system that measures industrial machine parameters and adjusts those industrial machine parameters in realtime to maximize industrial machine efficiency and to reduce or eliminate exceedances.

BACKGROUND

There are many negative aspects of the machining process which could impact Overall Equipment Efficiency (OEE). These include, but are not limited to, vibrations (“chatter”), worn spindle bearings, worn table bearings, poor thermal stability, the over-usage of processing coolant. Results include part quality (poor), machine life (reduced), or a machine efficiency (reduced).

There are numerous approaches to reducing chatter. For example, United States Patent Application Publication No. 20020146296 discloses that in milling operations, periodically sensed vibration signals synchronous with tool revolution enables a determination of whether the tool returns to approximately the same position each revolution. If so, stability is indicated by tightly grouped values of the periodically sensed vibration signal. If the tool does not return to the same position, spread in the value of the periodically sampled vibration signals is produced thereby indicating chatter conditions. Variance values may be calculated and displayed; histograms may be produced and displayed; corrective action, if needed, may be taken in response to the variance values and/or histogram. Nominal (or commanded) spindle speed, while not necessarily exactly synchronous with actual tool rotation, is entirely adequate to trigger samples and achieve clear indication of the presence or absence of chatter. This system is not autonomous and requires an operator.

United States Patent Application Publication No. 20160116899 discloses a system, method, and computer-readable medium for providing a user interface. The system includes circuitry configured to generate chatter information based on sensor data collected from a machining operation of a machine performed at a previously selected tool speed setting. The chatter information includes a chatter level value and a chatter frequency value. A plurality of different candidate tool speed settings is determined based on the generated chatter frequency value from the machining operation. The circuitry generates the user interface that includes a plurality of different tool speed settings, including the previously selected tool speed setting and the plurality of different candidate tool speed settings for selection by a user. The user interface is configured to indicate the chatter level value for the previously selected. This system requires an operator and is not autonomous.

United States Patent Application Publication No. 20100104388 discloses a vibration suppressing method and a vibration suppressing device. After a tool is attached to a main spindle, a modal parameter of the tool or a workpiece is computed. Thereafter, a relation between chatter frequency and phase difference is calculated as an approximation formula based on the obtained modal parameter and machining conditions. If chatter vibration occurs after initiation of the machining, a chatter frequency corresponding to a target phase difference is obtained using the approximation formula, and based on the obtained chatter frequency, the number of tool flutes and the main spindle rotation speed, the optimum rotation speed is calculated. The rotation speed of the main spindle is then changed in accordance with the obtained optimum rotation speed. There is no capability for machine learning.

Chinese Patent Application No. 104898565 discloses a chatter database system, which includes a central chatter database, which is fed with data corresponding to the machining and chatter conditions of machining tools, particularly a milling, turning, drilling or boring machine. The invention is characterized by the features that the data fed to the central chatter database is obtained and collected from at least two individual machining tools included in the chatter database system. Whereby the data is sent to the central chatter database via a data connection, preferably via a secured network, to generate chatter stability maps based on real encountered conditions. There is no capability for reducing chatter autonomously. As disclosed in Science Direct 11th CIRP Conference on Intelligent Computation in Manufacturing Engineering, CIRP ICME '17 “Online learning of stability lobe diagrams in milling” (Jens Friedricha, Jonas Torzewskia, Alexander Verla) the productivity of milling machines is limited by chatter vibrations. Stability lobe diagrams (SLD) allow the selection of suitable process parameters to maximize the productivity. However, the calculation of SLDs is very time-consuming and requires complex experiments. In this article a new online learning method is presented, which allows the calculation of SLDs during the production process. The algorithm is a combination of reinforcement learning and nearest-neighbor-classification and allows the estimation of the stability border based on measured vibration signals during machining. The proposed algorithm is capable of being continuously trained with sorted input data. A trust criterion is introduced, which allows judging the prediction quality of the algorithm. The algorithm is validated with analytical benchmark functions and with a 2-DOF milling stability simulation, maximizing the productivity. There exist several possibilities of generating SLDs. The SLD can be calculated based on the stability of the mathematical model with time delay. This demands a very accurate model. An experimental approach avoids the modelling by applying test cuts for different spindle speeds and cutting depths. The model-based as well as the experimental approach suffer from the additional effort of identifying the model or the SLD. Moreover, the SLD is only valid for the timeframe, in which the experiments were performed, as the machine behavior can change over time. In this paper, an approach to continuously learn the SLD during productive milling is presented. The learning algorithm allows the prediction of the stability border based on measured vibration signals. The application during productive milling leads to sorted input data and an incomplete training set. The algorithm can be trained with incomplete, sorted training data and the proposed trust criterion allows to judge the prediction reliability.

What is needed is a system for monitoring industrial machine tool parameters in real-time and correcting the processing conditions to maximize industrial machine efficiency. It would be preferable if sensor data were communicated wirelessly to a computing unit of the system. It would be further preferable if the system could autonomously alter the processing conditions of the industrial machine to maximize industrial machine efficiency in response to compromising processing conditions. It would be preferable if the system included a learning module. It would be preferable that the computing unit was configured to analyze the sensor data, archive the sensor data, archive the analyzed data and develop correction protocols using the analyzed data.

SUMMARY

The present technology is a system for monitoring industrial machine tool parameters in real-time and correcting the processing conditions to maximize industrial machine efficiency. The sensor data are communicated wirelessly to the computing unit of the system. The system can autonomously alter the processing conditions of the industrial machine to maximize industrial machine efficiency in response to compromising processing conditions. The system includes a learning module. In order to learn, the computing unit includes software configured to analyze the sensor data, archive the sensor data, archive the analyzed data and develop correction protocols using the analyzed data. Original Equipment Manufacturer (OEM) industrial machine data is also communicated to the software, including but not limited to g-code, m-code, spindle torque, and spindle current draw. OEM industrial machine data is then married to the sensor data to develop a more complete picture of the manufacturing process.

In one embodiment a machine-based data acquisition system (DAS) is provided for use with an industrial machine which includes an interior, an exterior and a control panel and is configured for use with a rotating tool, the machine-based DAS comprising an exterior sensor system which includes a sensor, an exterior sensor printed circuit board (PCB) and an exterior multi-sensor fusion and transmission (MSFT) module which includes an exterior MSFT PCB and a data transmitter; an exterior housing that is configured for placement on the exterior of the industrial machine and houses the exterior sensor system; a first interior sensor system which includes a first interior sensor, a first interior sensor PCB and a first interior MSFT module which includes a first interior MSFT PCB and a first interior data transmitter; a second interior sensor system which includes a second interior sensor, a second interior sensor PCB and a second interior MSFT module which includes a second interior MSFT PCB and a second interior data transmitter; a first interior housing that is configured for placement in the interior of the industrial machine and houses the first interior sensor system; a second interior housing that is configured for placement in the interior of the industrial machine and houses the second interior sensor system; a computing unit which includes a data analysis module; and a signal reception and integration module which includes a PCB and data reception and transmission connector and is configured to receive and process signals from the exterior MSFT module, the first interior MSFT module and the second MSFT module and to send data to the computing unit.

In the machine-based DAS, the exterior MSFT module, the first interior MSFT module, and second interior MSFT module may be configured to process signals from each sensor system to minimize signals arising from an ambient environment and to maximize signals arising from machine operations.

In the machine-based DAS, the signal reception and integration module may be configured to transmit a summary signal to the computing unit.

In the machine-based DAS, the data analysis module may be configured to detect features of the summary signal that are indicative of an exceedance.

In the machine-based DAS, the computing unit may be configured to send instructions to the control panel of the industrial machine.

In the machine-based DAS, the data analysis module may be configured to correlate operating conditions of the industrial machine with features of the summary signal and to develop a database.

In the machine-based DAS, the data analysis module may be configured to learn from the database to provide a learned database.

In the machine-based DAS, the data analysis module may be configured to reduce or eliminate the exceedance using the learned database.

In the machine-based DAS, the sensors may be acoustic sensors.

In another embodiment, a combination comprising the machine-based DAS and an industrial machine with a rotating tool is provided, the industrial machine including an exterior, an interior and a control panel, the machine-based DAS comprising: a signal reception and integration module which includes a wireless radio; an exterior sensor system which includes a sensor, and an exterior data transmission connector and is configured to detect signals and to transmit data to the signal reception and integration module; an exterior housing that is configured for placement on the exterior of the industrial machine and houses the exterior sensor system; a first interior sensor system which includes a first interior sensor and a first interior data transmission connector and is configured to detect signals and to transmit data to the signal reception and integration module; a second interior sensor system which includes a second interior sensor and a second interior data transmission connector and is configured to detect signals and to transmit data to the signal reception and integration module; a first interior housing that is configured for placement in the interior of the industrial machine and houses the first interior sensor system; a second interior housing that is configured for placement in the interior of the industrial machine and houses the second interior sensor system; and a computing unit which includes a data analysis module and is configured to receive and analyze data from the signal reception and integration module.

In the combination, each of the external sensor system, the first interior sensor system, and the second interior sensor system may include a multi-sensor fusion and transmission module which is in wired or wireless communication with the signal reception and integration module.

In the combination, the exterior sensor system, the first interior sensor system and the second interior sensor system may each include a pair of sensors.

In the combination, the sensors may be microelectromechanical acoustic sensors.

In another embodiment, a method of reducing or eliminating an exceedance in an industrial machine with a rotating tool is provided, the method comprising:

-selecting a machine-based DAS, the machine-based DAS including: an exterior sensor system which includes a sensor and a multi-sensor fusion and transmission module; an exterior housing that houses the exterior sensor system; a first interior sensor system and a second interior sensor system, each including a sensor and a multi-sensor fusion and transmission module; a first interior housing that houses the first interior sensor system; a second interior housing that houses the second interior sensor system; a computing unit which includes a data analysis module; and a signal reception and integration module which includes a wireless radio and is in communication with the computing unit and the multi-sensor fusion and transmission modules;

-attaching the exterior housing to an exterior surface of the industrial machine and attaching the first and the second interior housings to an interior surface of the industrial machine;

-processing a workpiece with the industrial machine and concomitantly, the machinebased DAS autonomously: sensing signals; processing the signals; and sending instructions to the control panel of the industrial machine to modify its operating parameters, thereby reducing or eliminating the exceedance.

In the method, the computing unit may send instructions to the industrial machine that reduce or eliminate the exceedance.

In the method, the instructions may be to modify a cut depth.

In the method, the instructions may be to modify a cut speed.

In the method, the exterior multi-sensor fusion and transmission module, the first interior multi-sensor fusion and transmission module and the second interior multi-sensor fusion and transmission module may process signals from each sensor system to minimize signals arising from an ambient environment and to maximize signals arising from processing a workpiece.

In the method, the signal reception and integration module may transmit a summary signal to the computing unit.

In the method, the data analysis module may detect features of the summary signal that are indicative of the exceedance.

In the method, the data analysis module may correlate operating conditions of the industrial machine with features of the summary signal that are indicative of the exceedance and develop a process stability database. In the method, the data analysis module may learn from the process stability database to provide a learned stability database.

In the method, the data analysis module may reduce or eliminate chatter using the learned stability database to determine instructions for the control panel of the industrial machine.

In the method, the signals may be acoustic signals.

FIGURES

Figure 1 is a perspective view of an exemplary system of the present technology with a partial cutaway to show the internal sensor systems in the work envelope.

Figure 2 is a block diagram of a system diagram for a machine with a rotating tool.

Figure 3 is a block diagram for an alternative embodiment of a machine with a rotating tool.

DESCRIPTION

Techniques operating according to the principles described herein may be implemented in any suitable manner. The processing and decision blocks of the flow charts above represent steps and acts that may be included in algorithms and/or circuitry that carry out these various processes. Unless otherwise indicated herein, the particular sequence of steps and/or acts described in each flow chart is merely illustrative of the algorithms that may be implemented and can be varied in implementations and embodiments of the principles described herein.

The acts performed as part of a method may be ordered in any suitable way. Accordingly, embodiments may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.

Except as otherwise expressly provided, the following rules of interpretation apply to this specification (written description and claims): (a) all words used herein shall be construed to be of such gender or number (singular or plural) as the circumstances require; (b) the singular terms "a", "an", and "the", as used in the specification and the appended claims include plural references unless the context clearly dictates otherwise; (c) the antecedent term "about" applied to a recited range or value denotes an approximation within the deviation in the range or value known or expected in the art from the measurements method; (d) the words "herein", "hereby", "hereof", "hereto", "hereinbefore", and "hereinafter", and words of similar import, refer to this specification in its entirety and not to any particular paragraph, claim or other subdivision, unless otherwise specified; (e) descriptive headings are for convenience only and shall not control or affect the meaning or construction of any part of the specification; and (f) "or" and "any" are not exclusive and "include" and "including" are not limiting. Further, the terms "comprising," "having," "including," and "containing" are to be construed as open-ended terms (i.e., meaning "including, but not limited to,") unless otherwise noted.

Recitation of ranges of values herein are merely intended to serve as a shorthand method of referring individually to each separate value falling within the range, unless otherwise indicated herein, and each separate value is incorporated into the specification as if it were individually recited herein. Where a specific range of values is provided, it is understood that each intervening value, to the tenth of the unit of the lower limit unless the context clearly dictates otherwise, between the upper and lower limit of that range and any other stated or intervening value in that stated range, is included therein. All smaller sub ranges are also included. The upper and lower limits of these smaller ranges are also included therein, subject to any specifically excluded limit in the stated range.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the relevant art. Although any methods and materials similar or equivalent to those described herein can also be used, the acceptable methods and materials are now described.

DEFINITIONS

Exceedance - in the context of the present technology, an exceedance includes, but is not limited to vibrations (“chatter”), worn spindle bearings, worn table bearings, poor thermal stability, the over-usage of processing coolant, results include part quality (poor), machine life (reduced), or a machine efficiency (reduced). Autonomous, non-interfering dynamic monitoring and correcting - in the context of the present technology, autonomous, non-interfering dynamic monitoring and correcting is effected by sensors and a printed circuit board that are integrated into the tool and operate without the assistance of a machine operator.

Computing unit - in the context of the present technology, a computing unit includes at least one processor, and computer-readable storage media. A computing unit may be, for example, a desktop or laptop personal computer, a personal digital assistant (PDA), a smart mobile phone, a server, a virtual graphics processing unit or any other suitable computing unit. A network adapter may be any suitable hardware and/or software to enable the computing unit to communicate wired and/or wirelessly with any other suitable computing unit over any suitable computing network. The computing network may include wireless access points, switches, routers, gateways, and/or other networking equipment as well as any suitable wired and/or wireless communication medium or media for exchanging data between two or more computers, including the Internet. Computer- readable media may be adapted to store data to be processed and/or instructions to be executed by processor. The processor enables processing of data and execution of instructions. The data and instructions may be stored on the computer-readable storage media (memory).

A computing unit may additionally have one or more components and peripherals, including input and output devices. Examples of output devices that can be used to provide a user interface include printers or display screens for visual presentation of output and speakers or other sound generating devices for audible presentation of output. Examples of input devices that can be used for a user interface include keyboards, and pointing devices, such as mice, touch pads, and digitizing tablets.

Communication network - in the context of the present technology a communication network includes but is not limited to a wireless fidelity (Wi-Fi [IEEE 802.11 ]) network, a light fidelity (Li-Fi) network, a satellite network, the internet, a cellular data network, a local area network (LAN), Bluetooth®, a wireless local area network (WLAN), or any combination thereof. The network adapter of the computing unit communicates via the communication network. Hammer Tap Test - in the context of the present technology, hammer tap testing or tap testing, also known as modal testing, is an experimental method that is used to excite the machine-tool system in order to extract its harmonic information such as natural frequencies, modal masses, modal damping ratios and mode shapes. This is normally done in static conditions, using the impact hammer as the excitation mechanism, and an accelerometer as the sensor. In theory, the tool tip should be given a perfect impulse which excites a range of frequencies with a constant amplitude in an infinitely short duration. This allows the user to obtain a clean frequency response function (FRF) over the full frequency range of interest.

Stability lobe diagrams (SLD) - stability lobe diagrams allow the selection of suitable process parameters to maximize the productivity.

Chatter - in the context of the present technology, is machine vibration and is generally a result of closed-loop, self-excitation. This closed-loop, self-excitation results from the varying chip thickness produced by out-of-phase oscillatory motion between a first and second cut path. The constructive amplitudes produced by out-of-phase oscillatory motion continue to increase dynamic motion within the CNC machine until a point of failure.

Sensor - in the context of the present technology, a sensor is a temperature sensor, an acoustic sensor or a vibration sensor. In each DAS the internal sensors and the externals sensor sense the same signal, for example, all three are temperature sensors.

Data transmitter - in the context of the present technology, a data transmitter is a wireless transmitter radio or a data transmission connector for wired communication. Wired communication may be over, for example, but not limited to, ethernet, universal serial bus, a Serial Advanced Technology Attachment, RS485, or RS232.

DETAILED DESCRIPTION

As shown in Figure 1 a machine-based DAS, generally referred to as 6 is for an industrial machine, generally referred to as 8. The machine-based DAS 6 includes a signal reception and integration module 10 that communicates with a control panel 12 of the industrial machine via a computing device 62 (see Figure 2) and three sensor systems, an external sensor system 14 and two internal sensor systems 16, 18. The external sensor system 14 is affixed to an exterior surface 20 of the industrial machine 8 and is configured to sense conditions, for example, temperature or sound or vibration. The two internal sensory systems 16, 18 are housed within the work envelope 22 and are configured to sense signals associated with processing. They are located on opposite sides of the work envelope 22 to one another. The industrial machine 8 may be a milling machine, or a turning machine and the tool 24 may be a rotating or stationary tool, respectively.

As shown in Figure 2, the sensor systems 14, 16, 18 each include a printed circuit board (PCB) 30 and a microelectromechanical sensor 32. The external and internal sensor systems 14, 16, 18 each further include a multi-sensor fusion and transmission module 26 which each include a wireless transmission radio 34, which may be a Bluetooth® radio, a power supply module 36 and a signal processor 38. In one embodiment the power supply module 36 is configured to retain a battery and includes a PCB 40 to monitor charge, discharge rate and other variables. In another embodiment, the power supply module 36 is a radio frequency wireless charging unit. Each sensor system 14, 16, 18, if they are acoustic sensor systems, is protected with an acoustically transparent but hydrophobic membrane such that an Ingress Protection Rating of 67 or greater is achieved, while not degrading the integrity of the acoustic signals. The internal sensor systems 16, 18 are each housed in a housing 42. The microelectromechanical sensors 32 are each in wired communication with the multi-sensor fusion and transmission module 26 via the PCB 30. The internal multi-sensor fusion and transmission modules 26 are further each in wireless communication with a signal reception and integration module 10. The external sensor system 14 is housed in a housing 48 on the exterior surface 20 of the industrial machine 8.

The signal reception and integration module 10 communicates with each sensor system and with the computing unit 62, which includes the data analysis module 50. The signal sent from the signal reception and integration module 10 is a summary signal of the three separate signals from the three sensor systems 14, 16, 18. The signal reception and integration module 10 is configured to process the signals from each sensor system 14, 16, 18 in order to minimize the signal arising from ambient signals in the environment and maximize the signal arising from machine operations and therefore includes a PCB 90. It is in wired communication with the computing unit 62. The data analysis module 50 communicates in wired fashion with the control panel 12.

The PCB 30 is independent from the signal processor 38 so that the sensor and PCB 30 can be situated in any orientation such to maximize signal reception without constraint from the multi-sensor fusion and transmission module 26, while the multi-sensor fusion and transmission module 26 may be situated in any orientation such to maximize Bluetooth transmission to the signal reception and integration module 10 (which includes a PCB).

In an alternative embodiment of the systems of Figure 2, the acoustic sensor systems 14, 16, 18 each include a pair of microelectromechanical acoustic sensors 32 to provide a directional microphone.

In an alternative embodiment of the systems of Figure 2, as shown in Figure 3, the external sensor system 14 includes the PCB 30, the microelectromechanical sensor 32, a power supply 70 and a multi-sensor fusion and transmission module 26, which may be a Bluetooth® radio. The power supply 70 is configured to retain a battery and includes a PCB 74 to monitor charge, discharge rate and other variables. In another embodiment, the power supply module 70 is a radio frequency wireless charging unit. The external sensor system 14 communicates wirelessly with the signal reception and integration module 10 and is housed in a housing 76 which is separate to the multi-sensor fusion module housing 78.

In an alternative embodiment of the systems of Figures 2 and 3, the external and internal sensor systems 14, 16, 18 may be hard wired to the signal reception and integration module 10 and therefore do not include the wireless radio 34 or the power supply 36.

In one embodiment, the sensors are vibration sensors. In another embodiment, the sensors are temperature sensors. In another alternative embodiment, the sensors are acoustic sensors. In all embodiments there may be more than two internal sensor systems and therefore, correspondingly more than two multi-sensor fusion and transmission modules 26.

In all embodiments, the first internal sensor system 16 is symmetrically opposite the second internal sensor system 18.

In an alternative to all embodiments, the wireless radios are replaced with wired connections. The multi-sensor fusion and transmission module 26, therefore comprises the signal processor (PCB) 38 and a data transmission connector for wired connection to the signal reception and integration module 10 and the signal reception and integration module 10, therefore comprises the PCB 90 and a data reception and transmission connector.

In an alternative to all embodiments, the data analysis module is a Web-based application.

In an alternative to all embodiments, the memory and processor are replaced with a system on a chip.

In an alternative to all embodiments, the wireless communication is via Bluetooth mesh networking.

In an alternative to all embodiments, the computing unit is a server which sends processed data or unprocessed data directly to the cloud. Cloud data is accessible by the same computing unit, a different computing unit comprising a web-based data analysis module or accessed by a mobile device.

Using acoustic sensor systems as an example, as noted above, the external acoustic sensor system 14 is for detecting and measuring ambient noise in the ambient environment. This is subtracted from the acoustic signals from the internal acoustic sensor system 16, 18 acoustic signals. The internal acoustic sensor systems 16, 18 detect and measure acoustic signals associated with processing and environmental acoustic signals within the work envelope 22. These are also subtracted from the acoustic signals associated with processing and include acoustic signals resulting from, for example, but not limited to coolant, spindle frequency, tool passing frequency, tooth passing frequency and swarf. The acoustic signals associated with processing may include spikes which are indicative of vibration (chatter). The redundancy in internal acoustic sensors 16, 18 allows for data sharing which will allow additional digital signal processing (DSP) to occur, for example, but not limited to overlap-add, overlap-subtract, or other methods to minimize noise, and maximize desirable signal extraction around the work envelop. Results from this will isolate actualized instances of undesirable features within a signal. Deploying this iterative check maximizes robustness and promotes processing optimization. The positioning of the first internal acoustic sensor system 16 and the second internal acoustic sensor system 18 maximizes collection and transmission of undesirable acoustic signals associated with processing and minimize collection and transmission of environmental acoustic signals in the work envelope 22. The mechanical structure of the acoustic sensor systems 14, 16, 18 ensures that no feature has a resonance between the frequency bandwidth range specified (dynamic stability) and is affixed by any means to cause unnecessary contamination of the acoustic signal through self-excitation.

The computing unit 62 comprising the data analysis module 50 includes a signal processing algorithm, a machine learning algorithm, and a self-learning algorithm.

The signal processing algorithm was developed to optimize OEE by modifying those relevant process parameters.

The signal processing algorithm receives an acoustic signal from the work envelope via the signal reception and integration module 10. The acoustic signal described the current state of the CNC machine system 8. The acoustic signal is analyzed by the data analysis module 50 within the computing unit 62. The data analysis module 50 compares the received acoustic signal with a pre-determined threshold. This threshold is established by a pre-determined acceptance criterion. If the acoustic signal exceeds the pre-determined threshold those relevant process parameters are modified.

When unwanted chatter occurs, the signal processing algorithm leverages the thresholds defined by the SLD for the current state of the CNC machine system 8 to modify the depth of cut and spindle speed according to those prescribed variation techniques. Ultimately, chatter is reduced or eliminated. The machine learning algorithm was developed to reduce or eliminate the need for the hammer tap test for CNC machine systems 8. From the broad and incomplete stability database created via the hammer tap test, the machine learning algorithm does establish a learned stability database from the acoustic signals received from the signal reception and integration module 10. The machine learning algorithm achieves this by correlating those relevant process parameters, constructing the SLD signature, and establishing a connection therebetween.

The machine learning algorithm receives an acoustic signal from the work envelope via the signal reception and integration module 10. The acoustic signal described the current state of the CNC machine 8. The acoustic signal is then analyzed by the data analysis module 50 within the computing unit 62. The data analysis module 50 compares the received acoustic signal with a hammer tap test stability database. Upon comparison, indictive features of the acoustic signal may be isolated and a SLD may be developed for the current state. The developed SLD and the corresponding CNC machine parameters may be stored in a learned stability databased for immediate or future exploitation.

The self-learning algorithm expands on the learned stability database, by automatically developing a self-learned stability database. Based on the learned stability database developed by the machine learning algorithm, the self-learning algorithm will capitalize on the SLD and corresponding CNC machine parameters to expand the learned stability database.

The following steps were taken by the signal processing algorithm in order to detect chatter. This process is brute force in execution, while other methods of execution using learning may be described and applied in other embodiments.

Further, this brute force method deploys only one DSP technique, that being amplitude, while in alternative embodiments other DSP techniques, such as density, root means square, kurtosis, or skewness, may be used in any combination.

1. The machine was operated and time domain acoustic signals were sent to the computing device as described above. 2. The time-domain acoustic signals were transformed using the Fast Fourier Transform (FFT) to generate an FFT spectrum.

3. The FFT spectrum was filtered for frequencies and magnitudes that were not spindle frequencies or magnitudes, tool passing frequencies or magnitudes, tooth passing frequencies or magnitudes or harmonics thereof in order to detect and identify the chatter frequencies and/or magnitudes.

4. If the filtered FFT magnitude was greater than largest unfiltered FFT magnitude the process was unstable and chatter magnitudes were detected. If the filtered FFT magnitude was less than or equal to the largest unfiltered FFT magnitude the process was stable. Similarly, if the filtered FFT frequency was greater than largest unfiltered FFT frequency the process was unstable and chatter frequencies were detected. If the filtered FFT frequency was less than or equal to the largest unfiltered FFT frequency the process was stable.

5. The chatter frequency and magnitude were isolated from the FFT spectrum and were displayed.

6. The machine feed was stopped and the SLD was used to determine which of discrete spindle speed tuning, discrete spindle depth tuning, discrete spindle width tuning, and/or continuous spindle speed variation was needed to reduce or eliminate the chatter.

7. The data were sent to the process stability database for use by the signal processing algorithm, the machine learning algorithm and/or the self-learning algorithm.

8. The computing unit 62 communicated in wired fashion with the control panel 12 and the control panel, under control of the data analysis module 10, automatically employed one or more of the following:

Information on those variation techniques used by the signal processing algorithm described earlier are provided. These variation techniques are specific for reducing or eliminating chatter. Other techniques may be used by the signal processing algorithm to reduce or eliminate other exceedances observed during the manufacturing process. Spindle Speed Variation: modified the spindle speed to be at the average bandwidth at lowest lobe order while remaining within the upper limit of the workpiece, machine and cutting tool boundaries. If spindle speed was outside of the upper limit of the workpiece, machine and cutting tool boundaries, the spindle speed was lowered to the average bandwidth at iteratively higher lobe orders, remaining within the lower limit of the workpiece, machine and cutting tool boundaries. If neither was effective at reducing or eliminating chatter, the continuous spindle speed variation sub-routine was employed. In situations where that was not effective at reducing or eliminating vibration, the depth of cut was automatically modified. If neither of these were effective, both the depth of cut and spindle speed were automatically modified.

Depth of Cut Variation: modified the depth of cut by automatically increasing or decreasing the depth of cut remaining at the current spindle speed to reach the stable- unstable intersection of the SLD and override the feed rate proportionally to maintain constant surface speed.

Spindle Speed and Depth of Cut Variation: modified the spindle speed and the depth of cut by automatically increasing the spindle speed to be at the average bandwidth at lowest lobe order while remaining within the upper limit of the workpiece, machine and cutting tool boundaries and automatically increase the current depth of cut to reach the stable- unstable intersection of the SLD. If spindle speed was outside of the upper limit of the workpiece, machine and cutting tool boundaries, the spindle speed was lowered to the average bandwidth at iteratively higher lobe orders, remaining within the lower limit of the workpiece, machine and cutting tool boundaries and automatically increase depth of cut. If increasing or decreasing the spindle speed was not effective, the depth of cut was automatically decreased to reach the stable-unstable intersection of the stability lobe diagram. The feed rate was over-ridden as needed to maintain constant surface speed.

Continuous Spindle Speed Variation: modify the spindle speed around some nominal command value, given those constraints provided by the machine, material, and cutting tool. The varying spindle speed therefore produces varying harmonic frequency. As a result, the varying frequency does not facilitate constructive wave superposition, and naturally, does not facilitate an increasing dynamic amplitude within the CNC machine. This process results in chatter mitigation.

The foregoing allows for autonomous, non-interfering dynamic monitoring and correcting to reduce or eliminate chatter.

While example embodiments have been described in connection with what is presently considered to be an example of a possible most practical and/or suitable embodiment, it is to be understood that the descriptions are not to be limited to the disclosed embodiments, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the example embodiment. Those skilled in the art will recognize or be able to ascertain using no more than routine experimentation, many equivalents to the specific example embodiments specifically described herein. Such equivalents are intended to be encompassed in the scope of the claims, if appended hereto or subsequently filed.