Numerical control machine tool automation collaborative control method and system

Abstract

The application discloses a numerical control machine tool automation collaborative control method and system, which takes into account solving the problems of heterogeneous numerical control machine tool collaborative processing signal normalization and coordinate unification, and improves the defects of bagging fruit easy to be damaged, process timing fixed, and unstable closing and sealing. The method comprises the following steps: S1, multi-dimensional sensing and collecting machine tool spindle load, feeding shaft position, visual coordinates and other multi-source original sensing data streams; S2, completing heterogeneous signal normalization by means of a protocol conversion matrix to obtain standardized working condition signals, segmenting and checking numerical values to remove noise, and outputting effective working condition data streams; S3, unifying local coordinates of each machine tool through a global coordinate mapping model to generate a global coordinate view signal; S4, dynamically updating machine tool process timing by adopting a timing weighted iterative sorting algorithm; and S5, superimposing timing offset compensation with a unified timestamp to generate aligned equipment motion instructions and issue the instructions, so that multi-machine tool synchronous collaborative processing is realized, three-level layered transmission channels and distributed timing storage are set, and data isolation transmission and multi-version timing backtracking calling are ensured.

Description

Technical Field

[0001] This invention relates to the field of multi-machine collaborative automation control technology for CNC machine tools, and specifically to an automated collaborative control method and system for CNC machine tools. Background Technology

[0002] In intelligent production lines, multiple CNC machine tools, along with auxiliary devices such as robotic arms and tool magazines, often collaborate to process complex parts. These devices frequently come from different manufacturers, and their controllers use incompatible proprietary communication protocols, data formats, and signal definitions. Multi-source sensor data, such as spindle load, feed position, and tool magazine status, are difficult to directly aggregate and uniformly analyze. System integration requires developing protocol conversion interfaces for each machine, resulting in long development cycles, poor scalability, and conversion delays that impact real-time performance. Furthermore, each machine tool in the workshop establishes a local coordinate system based on its own mechanical origin. After workpieces move between machines, they need to be re-clamped, tool-set, and their local coordinates calibrated. Repeated positioning not only increases auxiliary time but also introduces cumulative errors, leading to inconsistent positional references in multi-machine collaborative processing and affecting overall assembly accuracy. Regarding process scheduling, existing scheduling systems mostly generate static plans based on preset fixed timeframes, failing to integrate real-time operating condition information such as spindle wear degradation, tool life decay, and raw material size deviations. Actual processing time varies dynamically, easily causing machine tool waiting or processing interference, severely reducing production line cycle time and equipment utilization. Furthermore, multi-machine collaboration places extremely high demands on time synchronization. However, conventional Ethernet-based clock synchronization accuracy is only at the millisecond level and lacks precise timing offset compensation for specific processes. Control commands from different devices cannot be strictly aligned on a unified time axis, easily leading to overlapping actions or timing misalignments and posing a collision risk. In terms of data transmission, raw sensor streams, intermediate calculation results, and final control commands often share the same channel, lacking hierarchical isolation. High-bandwidth signals easily crowd out command bandwidth, causing control jitter and compromising determinism. These problems—difficulty in heterogeneous fusion, inconsistent coordinates, dynamic timing mismatch, low synchronization accuracy, and data stream conflicts—severely restrict the automation level and flexible production capabilities of multi-machine collaborative processing. There is an urgent need for a collaborative control method that supports signal normalization, global coordinate mapping, real-time process iterative optimization, and highly reliable hierarchical transmission. Summary of the Invention

[0003] In view of the shortcomings of the prior art, the purpose of this invention is to provide an automated collaborative control method and system for CNC machine tools, which solves the problem of the inability to achieve signal normalization and coordinate unification for collaborative operation of heterogeneous CNC machine tools. This invention utilizes multi-dimensional sensing acquisition, employs a protocol conversion matrix to normalize heterogeneous signals, and performs segmented verification and noise reduction; establishes a global coordinate mapping to unify the coordinates of each machine tool; optimizes the process timing based on weighted iterative sorting of spindle loss and coordinate deviation, and superimposes timing offset compensation to generate time-aligned sub-device motion commands, which are then distributed through a three-level isolated transmission channel to achieve synchronous collaborative machining of multiple machines.

[0004] This invention provides an automated collaborative control method for CNC machine tools, comprising: S1: Collect spindle load signals, feed axis position signals, tool magazine positioning signals, workpiece visual coordinate signals, and motion status signals of auxiliary devices from multiple CNC machine tools through multi-dimensional sensing units to form a multi-source raw sensor data stream; S2: Using multi-source raw sensor data streams, the signal normalization and conversion of heterogeneous devices is completed through a preset protocol conversion matrix operation model to form standardized operating condition numerical signals; noise interference signals are removed through a segmented numerical verification algorithm to form an effective operating condition data stream; S3: Using effective working condition data stream, the conversion from local coordinates of each machine tool to global coordinates is completed through the global workpiece coordinate mapping calculation model, forming a global coordinate view signal; S4: Using effective working condition data streams and global coordinate view signals, generate updated process execution timing signals for each machine tool through a real-time process timing weighted iterative sorting algorithm; S5: Use the updated process execution timing signal, add timing offset compensation based on a unified timestamp, form time-aligned sub-equipment motion execution command signals, and output them to each CNC machine tool and auxiliary device to complete synchronous collaborative processing.

[0005] In one embodiment of the present invention, after acquiring multi-source raw sensor data streams, a dedicated fixed data frame header identifier is configured for each type of sensor signal. The multi-source raw sensor data streams carrying different header identifiers are divided into three independent data transmission channels to complete layered transmission. The first-level channel only transmits multi-source raw sensor data streams without conversion processing, the second-level channel only transmits standardized chemical condition numerical signals after normalization conversion, and the third-level channel only stores various timing signals and motion execution command signals to be sent. The three types of data transmission channels are configured with independent transmission buffer spaces that are not interchangeable. Various types of messages can only be transferred within the corresponding level channel. Only one-way data message transfer operations are allowed between different level channels. The corresponding level of operation is only performed after the message header identifier is identified at the cross-level transfer node. Throughout the process, different types of data streams are not mixed for storage and synchronous parsing operations in the same transmission buffer space.

[0006] In one embodiment of the present invention, before performing the heterogeneous device signal normalization conversion operation, complete conversion matrix groups corresponding to different types of CNC machine tool controllers and their supporting auxiliary device controllers are pre-entered. The conversion matrix group contains independent conversion matrix units adapted to the native signal protocols of various devices. When retrieving multi-source raw sensor data streams, the conversion matrix units under the corresponding device type are matched according to the frame header identifier carried by the data stream. All values ​​in a single frame of raw sensor data stream are sequentially substituted into the matched conversion matrix units to complete the unified bit format conversion. After all sensor values ​​in a single message have completed the matrix conversion, they are uniformly encapsulated to generate a complete standardized operating condition numerical signal message. All message encapsulation operations are performed according to a unified fixed message length standard. All converted standardized operating condition numerical signals are stored in the dedicated buffer space corresponding to the secondary data transmission channel for subsequent verification and processing. The bit format conversion calculation process is as follows:

[0007] in, The normalized bit output value of the i-th sensor signal. Let be the bit scaling factor for the protocol of the k-th device. Let the coefficient of the transformation matrix unit of the k-th type of equipment be the matrix coefficient in the i-th row and j-th column. For the j-th channel of the single-frame raw data stream, is the fixed offset of the native protocol for the k-th type of device, and n is the total number of sensor signals contained in a single frame message.

[0008] In one embodiment of the present invention, before performing the global workpiece coordinate mapping calculation model operation, the coordinate value set corresponding to the fixed reference origin in the workshop is pre-entered. The local workpiece coordinate values ​​of a single machine tool and the coordinate values ​​captured by the transfer auxiliary device contained in the effective working condition data stream are retrieved. The two sets of coordinate values ​​are synchronously input into the global workpiece coordinate mapping calculation model. The model sequentially performs coordinate translation operation and coordinate rotation conversion operation to complete the numerical conversion from local coordinates to the unified global coordinates of the workshop. After all workpiece coordinate values ​​of a single device are converted, they are integrated into a single device global coordinate value set. After the global coordinate value sets corresponding to multiple devices are summarized and integrated, a complete global coordinate view signal is generated. The generated global coordinate view signal is separately encapsulated with a dedicated message header identifier and stored in the secondary data transmission channel cache space for subsequent process timing operation calls. The calculation formula for the global coordinate numerical conversion is as follows:

[0009] in, Let p be the global reference coordinate value of the workpiece on the p-th device. For three-axis standardized sensor coordinate values, This refers to the rotation angle of the equipment station relative to the workshop reference origin. , , This is the three-axis translation compensation amount for the workshop reference origin.

[0010] In one embodiment of the present invention, after generating the global coordinate view signal, the complete global coordinate view signal is split into independent coordinate sub-signals for each device and stored in mutually isolated partitioned buffer units within the secondary data transmission channel. When receiving a signal retrieval request from the process timing calculation unit, the device coordinate sub-signals stored in the corresponding partitioned buffer unit are extracted according to the device identifier carried in the request. The extracted coordinate sub-signals are synchronously transmitted to the process timing weighted iterative sorting algorithm calculation unit along with the valid working condition data stream of the same batch. All coordinate sub-signal retrieval operations can only be executed sequentially according to the processing batch order. The coordinate signal extraction and transmission operation of the next processing batch is not started before all coordinate sub-signals of the current batch are retrieved. During the transmission process, the original complete message of the global coordinate view signal is continuously retained in the buffer unit without being overwritten or deleted.

[0011] In one embodiment of the present invention, when the real-time process timing weighted iterative sorting algorithm is started, the set of values ​​corresponding to spindle loss and the set of values ​​corresponding to workpiece incoming coordinate deviation are first extracted from the effective working condition data stream. Independent weighted correction coefficient sets are configured for the two sets of values ​​respectively. The two sets of values ​​and the corresponding weighted correction coefficient sets are synchronously input into the real-time process timing weighted iterative sorting algorithm to perform a single timing iteration operation. During the single iteration operation, the original standard execution time value of a single process is first corrected according to the weighted correction coefficients. After all process time values ​​are corrected, the execution order of the processes corresponding to each machine tool is rearranged. After the order is arranged, the corrected process time values ​​of all machine tools are integrated to generate a complete updated process execution timing signal for each machine tool. The process execution timing signal generated by the single iteration operation is uniformly added with a batch identifier and stored in the three-level data transmission channel buffer unit. The calculation process of the single timing iteration operation is as follows:

[0012] in, The execution time of the m-th process is adjusted after the t-th iteration. The spindle loss weighting factor. This represents the real-time spindle wear value for the m-th process. This is the weighting coefficient for coordinate deviation. Let m be the global coordinates of the workpiece. Here, t represents the standard global coordinates of the workpiece, and t represents the current iteration round. M represents the original standard duration of the process, and M represents the total number of processes in a single batch.

[0013] In one embodiment of the present invention, after generating the updated process execution timing signals for each machine tool, multiple independent timing signal cache partitions are divided according to the equipment type. The process execution timing signals corresponding to different machine tools are stored in their respective dedicated cache partitions. Each iteration operation only updates the timing signal value set corresponding to a single batch of processing, while the timing signal value sets of other historical batches remain in their original storage state without being overwritten or modified. When starting the next round of iteration operation, only the process execution timing signal corresponding to the current batch to be processed, the corresponding valid working condition data stream, and the global coordinate view signal are read. The multiple versions of process execution timing signals generated by multiple rounds of iteration operation are stored in layers according to the iteration generation order. When retrieving timing signals, the timing signal message generated by any round of iteration operation can be extracted based on the iteration batch identifier.

[0014] In one embodiment of the present invention, when performing segmented numerical verification algorithm processing on the standardized operating condition numerical signal, all sensor values ​​within the standardized operating condition numerical signal message are extracted frame by frame. A set of pre-configured numerical ranges corresponding to various types of sensor values ​​is retrieved. Each sensor value in a single frame message is compared with the corresponding numerical range set in turn. After comparison, sensor values ​​that are within the corresponding range are retained, and interference sensor values ​​that exceed the range are removed. All retained valid sensor values ​​are re-integrated and encapsulated to form a complete single-frame valid operating condition data stream message. All messages that have completed the verification operation are uniformly summarized and integrated to generate a complete valid operating condition data stream overall message and transmitted to the secondary data transmission channel to await coordinate conversion calculation. During the verification process, the original standardized operating condition numerical signal message is completely retained without deletion.

[0015] In one embodiment of the present invention, after generating the updated process execution timing signal, the equipment is divided into independent equipment groups according to all CNC machine tools and auxiliary devices in the workshop. A dedicated timing offset compensation value set is assigned to each equipment group. After retrieving the process execution timing signal of the corresponding equipment group, the total duration value of the timing signal is superimposed with the dedicated timing offset compensation value set of the group. After the superposition operation is completed, a sub-equipment motion execution instruction sub-signal is generated to adapt to the single equipment group. After the motion execution instruction sub-signals corresponding to all equipment groups are integrated to form a complete motion execution instruction signal, it is sent to the corresponding equipment controller according to the independent buffer partition of the three-level data transmission channel. The instruction sub-signal sending links of different equipment groups are isolated from each other and there is no message cross-transmission operation process.

[0016] The present invention also provides an automated collaborative control system for CNC machine tools, comprising: The multi-dimensional signal acquisition module collects various sensor signals from machine tools and auxiliary devices and outputs multi-source raw sensor data streams. The layered data stream transmission module is equipped with three independent transmission channels that respectively carry the raw sensor data stream, standardized operating condition numerical signals, and timing command signals. The computation and processing module integrates a protocol conversion unit, a numerical verification unit, a coordinate mapping unit, and a timing iteration unit. It receives the output messages from the layered data stream transmission module and sequentially generates standardized working condition numerical signals, effective working condition data streams, global coordinate view signals, and process execution timing signals. The timing instruction generation module retrieves the process execution timing signal, superimposes the timing offset compensation value to obtain the sub-equipment motion execution instruction signal, and outputs it to the external processing equipment. The distributed time-series storage module binds batch time sequence numbers to various signals to complete local storage of fragments. Upon receiving a retrieval request, it integrates the fragments according to the number to restore the complete signal message. The five modules independently complete the corresponding signal processing operations. Only standardized signals at the matching level are transmitted between modules, and the transmission links of each module are isolated from each other.

[0017] This invention provides an automated collaborative control method and system for CNC machine tools. It acquires data through multi-dimensional sensing, normalizes heterogeneous signals using a protocol conversion matrix, and performs segmented verification and noise reduction. It establishes a global coordinate mapping to unify the coordinates of each machine tool. Based on spindle loss and coordinate deviation, it optimizes the process timing by weighted iterative sorting and superimposed timing offset compensation to generate time-aligned sub-equipment motion commands, which are then distributed through a three-level isolated transmission channel to achieve synchronous collaborative machining of multiple machines. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of a method for automated collaborative control of CNC machine tools; Figure 2 A flowchart for multi-source sensing layered transmission and standardized verification; Figure 3 A flowchart for global workpiece coordinate mapping and batch coordinate retrieval loop; Figure 4 Flowchart for real-time process sequence weighted iterative loop and equipment instruction issuance Figure 5 This is a system architecture diagram of an automated collaborative control system for CNC machine tools. Detailed Implementation

[0020] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0021] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0022] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0023] Please see Figure 1-5 The figure shows an automated collaborative control method and system for CNC machine tools according to the present invention. The present invention provides an automated collaborative control method for CNC machine tools, comprising: S1: acquiring spindle load signals, feed axis position signals, tool magazine positioning signals, workpiece visual coordinate signals, and motion status signals of auxiliary devices from multiple CNC machine tools through multi-dimensional sensing units to form a multi-source raw sensor data stream; S2: using the multi-source raw sensor data stream, completing the normalization conversion of heterogeneous equipment signals through a preset protocol conversion matrix operation model to form a standardized working condition numerical signal; and eliminating noise interference signals through a segmented numerical verification algorithm to form an effective working condition data stream; S3: using the effective working condition data stream, completing the conversion from local coordinates of each machine tool to global coordinates through a global workpiece coordinate mapping calculation model to form a global coordinate view signal; S4: using the effective working condition data stream and the global coordinate view signal, generating updated process execution timing signals for each machine tool through a real-time process timing weighted iterative sorting algorithm; S5: using the updated process execution timing signals, adding timing offset compensation based on a unified timestamp to form time-aligned sub-equipment motion execution command signals, and outputting them to each CNC machine tool and auxiliary device to complete synchronous collaborative machining.

[0024] like Figure 1As shown, the system's perception layer consists of multi-dimensional sensing units distributed across multiple CNC machine tools and auxiliary devices. These units are responsible for capturing the most primitive physical state of the production site in real time. The types of signals collected are extremely rich, covering spindle load signals reflecting cutting load, feed axis position signals describing the relative position of the tool and workpiece, tool magazine positioning signals ensuring correct tool preparation, workpiece visual coordinate signals used to replace traditional manual tool setting and clamping, and motion status signals from auxiliary devices such as robotic arms and rotary tables. These signals from different sources and in different formats converge to form a multi-source raw sensor data stream, providing a comprehensive data foundation for the intelligent decision-making of the entire system.

[0025] To address the data congestion issues inherent in traditional integration methods, the system employs a rigorous three-tiered data transmission architecture. Upon entering the transmission channel, each type of sensor signal is assigned a unique, fixed data frame header identifier, serving as its "identity card" throughout its lifecycle. Subsequently, all data is strictly divided into three physically or logically independent transmission channels. The first-level channel acts as a "fast track" for raw data, allowing only the raw sensor data stream carrying the header identifier but without any processing to pass through, ensuring the data's integrity and real-time performance. The second-level channel is the core processing area, specifically carrying standardized chemical condition numerical signals after normalization. The third-level channel is a storage area used to store various timing signals and motion execution commands to be issued. These three channels each possess independent, non-interoperable transmission buffer spaces, fundamentally preventing the mixed storage and synchronous parsing of different types of messages within the same buffer, thereby eliminating data flow conflicts. Inter-channel interaction is strictly limited to unidirectional flow. Only when a message needs to be processed or sent across layers will the system identify the message header identifier at the flow node, capture it, and send it to the corresponding layer's processing unit for processing. After processing, it will be encapsulated and sent to the next level channel.

[0026] like Figure 2As shown in the flowchart, this flowchart fully demonstrates the end-to-end processing logic of multi-source sensor layered transmission and standardized verification. The starting node of the process is the multi-dimensional sensor unit that collects multi-source raw data streams. This node is responsible for synchronously collecting various sensor information such as spindle load, feed axis position, tool magazine positioning, workpiece visual coordinates, and motion status of all CNC machine tools in the workshop. It integrates the raw values ​​collected by all devices to form a unified multi-source raw sensor data stream, which is the source of all data in the entire control process. After the collection is completed, it enters the process of adding frame header identifiers and diverting the data stream to the three-level independent transmission channel node. This node will assign a dedicated fixed frame header identifier for each type of sensor signal, and use the identifier to distinguish the signal attributes and divert the data stream to the three completely independent transmission channels. The three types of channels have dedicated buffer spaces that are not interchangeable. The data stream can only flow unidirectionally across levels, and there will be no situation where different types of data are mixed for storage and parsing. The subsequent channel branch judgment node undertakes the data diversion and filtering function. Based on the frame header identifier inherent in the message, it distinguishes the data processing direction. One branch flows to the primary channel, storing the raw sensor data stream. This channel only stores the raw acquisition messages without any conversion processing, without performing any numerical calculations; it only performs raw data backup. The other branch flows to the secondary channel, performing heterogeneous device signal normalization conversion. After entering the secondary channel, it activates the core operation node for heterogeneous device signal normalization conversion matrix operation. This node calls the pre-stored conversion matrix units specific to various devices, substituting all values ​​in the single frame of raw data stream into the matrix to complete the unified bit format conversion. After all values ​​are converted, the output standardized operating condition numerical signal is stored in the secondary buffer node. The standardized numerical messages are uniformly encapsulated according to a fixed length and stored in the secondary channel's dedicated buffer for subsequent verification processing. The bit format conversion calculation process is as follows:

[0027] in, The normalized bit output value of the i-th sensor signal. Let be the bit scaling factor for the protocol of the k-th device. Let the coefficient of the transformation matrix unit of the k-th type of equipment be the matrix coefficient in the i-th row and j-th column. For the j-th channel of the single-frame raw data stream, Here, n represents the fixed offset of the native protocol for the k-th type of device, and n is the total number of sensor signals contained in a single frame. The heterogeneous device signal normalization conversion formula is the core computational carrier for achieving unified quantization processing of raw sensor data from multiple machine tools and auxiliary devices. The calculation relies on pre-stored conversion matrix groups specific to various devices, adapting to the objective reality of inconsistent bit widths, numerical ranges, and zero-point references in the native signal protocols of different manufacturers' equipment in the workshop. (The formula contains...) The normalized bit output value of the i-th sensor signal. Let be the bit scaling factor for the protocol of the k-th device. Let the coefficient of the transformation matrix unit of the k-th type of equipment be the matrix coefficient in the i-th row and j-th column. For the j-th channel of the single-frame raw data stream, Let n be the fixed offset of the native protocol for the k-th type of device, and n be the total number of sensor signals contained in a single frame. This method unifies the scaling of raw numerical ranges across different devices, eliminating magnitude gaps caused by differences in hardware sampling intervals. Matrix multiplication and summation operators work together to traverse all sensor values ​​in a single frame, using matched transformation matrix units to complete linear mapping and adapt to device protocols. Addition with a fixed offset compensates for inherent zero-point errors from the device's manufacturing process. 0.8-power nonlinear operations mitigate signal drift interference caused by long-term machine tool vibration and sensor temperature rise. This operation is performed only at the cross-level transfer node from the primary raw data flow to the secondary channel. The standardized values ​​after conversion are uniformly encapsulated into fixed-length messages and stored in a dedicated secondary buffer, providing a unified dimensional input for the segmented numerical verification algorithm. Without this formula, the units and references of raw signals such as spindle load and visual coordinates would be inconsistent, leading to significant offsets in subsequent coordinate conversions. Data within the layered transmission channel cannot be parsed correctly, and the underlying data consistency of multi-device collaboration cannot be guaranteed. The entire preprocessing link would generate a large amount of distorted data, directly interfering with subsequent coordinate mapping and time-series iteration operations.

[0028] After the cache is stored, a segmented numerical verification algorithm is used to filter noise nodes frame by frame. This node extracts all sensor values ​​within the message frame by frame, retrieves the preset valid range of various sensor values, and compares and verifies each value one by one. The comparison results are sent to the numerical range verification branch node to complete data diversion and filtering. Distorted interference data that exceeds the preset numerical range will be directly processed by the out-of-range noise signal discard node. Normal sensor values ​​within the valid range will be sent to the valid value encapsulation valid working condition data stream node. The filtered valid values ​​are re-integrated and encapsulated into a complete single-frame message. After all the verified messages are summarized and integrated, a complete valid working condition data stream is generated. Finally, the valid working condition data stream is sent to the coordinate mapping unit node to provide accurate and noise-free standardized input data for subsequent unified global coordinate conversion in the workshop. The entire process relies on three layers of processing logic—layered channel isolation, matrix standardization conversion, and segmented numerical verification—to filter out distorted data layer by layer and unify the signal dimensions of equipment. Each process node performs its own function and works in an orderly manner, forming a complete closed-loop preprocessing link from raw data acquisition to effective working condition data output. This completely solves the fundamental technical problems of inconsistent sensor signal formats and large noise interference from different models of machine tools and auxiliary devices.

[0029] like Figure 3As shown in the flowchart, this flowchart presents the complete process of global workpiece coordinate mapping and batch coordinate retrieval loop processing. The process starts at the node for receiving valid working condition data stream. This node receives valid working condition data after noise filtering and standardization conversion, retaining only the normal sensor values ​​within the range, thus avoiding the calculation deviation caused by distorted coordinate values ​​to the global coordinate conversion from the root. After the data reception is completed, the node for loading the workshop reference origin coordinate set enters. This node preloads the complete set of coordinate values ​​corresponding to the unified fixed reference origin of the workshop, providing a unified spatial reference standard for the local coordinate conversion of all equipment, eliminating the inherent defect of inconsistent references of the independent coordinate systems of each machine tool. After the reference coordinate parameters are loaded, the core operation node for coordinate rotation and translation calculations in the global workpiece coordinate mapping model is entered. This node synchronously reads two sets of coordinate data from the effective working condition data stream: the local workpiece coordinates of a single machine tool and the coordinates captured by the transfer auxiliary device. It then sequentially performs two layers of numerical conversion operations: coordinate translation and coordinate rotation, compensating for spatial coordinate differences caused by equipment installation offsets and workstation rotations. After all workpiece coordinates for a single machine are converted, they are integrated to form an independent coordinate set for that machine. After the coordinate sets from multiple machines are aggregated, the node generates a complete global coordinate view signal, integrating all unified reference coordinates and encapsulating them with a unique message header identifier. All data is stored in the secondary transmission channel buffer. The calculation formula for the global coordinate numerical conversion is as follows:

[0030] in, Let p be the global reference coordinate value of the workpiece on the p-th device. For three-axis standardized sensor coordinate values, This refers to the rotation angle of the equipment station relative to the workshop reference origin. , , This provides the three-axis translation compensation for the workshop's reference origin. Two sets of summation operators integrate the standardized three-axis coordinate inputs. Cosine and sine trigonometric functions complete the coordinate rotation transformation corresponding to the workstation rotation angle. The square operator amplifies the coordinate offset difference to facilitate the identification of incoming material positioning errors. The square root operator calculates the planar Euclidean distance to quantify the planar offset. Linear addition is used for superposition. Correcting Z-axis height installation deviation. Before calculation, the unified reference origin coordinate set of the workshop is pre-loaded. The local workpiece coordinates of the machine tool and the coordinates captured by the transfer device are read synchronously and substituted into the formula. After conversion, the global coordinate set of a single device is integrated, and all equipment data is summarized to generate a global coordinate view signal, which is stored in a secondary channel isolation cache. It is then split into independent coordinate sub-signals and stored in partitions. The signals are retrieved strictly according to the processing batch order for use by the timing calculation unit. The absolute value of the difference between the global coordinates generated by the formula calculation and the standard coordinates is the core deviation input parameter of the timing iterative algorithm, which connects the underlying standardized data with the upper-level timing optimization logic. Without this formula, there is no unified reference standard for the coordinates of each machine tool, the workpiece incoming offset cannot be quantified, the process timing arrangement can only use ideal standard coordinates, and in actual processing, collaborative failures such as workpiece positioning offset, process timeout, and equipment idling congestion will frequently occur.

[0031] After the coordinate view signal is generated, it is split into independent coordinate sub-signals for each device and stored in an isolated partition cache node. The integrated complete global coordinate view is split into independent coordinate sub-signals corresponding to each device. The sub-signals of different devices are stored in mutually isolated partition cache units within the secondary channel. The original complete global coordinate view message is continuously retained in the cache unit throughout the process and will not be overwritten or deleted by newly written data. After the cache storage is completed, it enters the batch retrieval loop judgment node. This node controls the retrieval order of coordinate sub-signals and is the core control node of the loop logic in the process. If the coordinate data of the current processing batch has not been fully retrieved, it enters the current batch not finished processing and loops to read the partition sub-signal node, continuously and repeatedly retrieving the coordinate sub-signals in the corresponding partition cache unit of the current batch until all device coordinate sub-signals of the current batch have been extracted. It will not start the coordinate reading and transmission operation of the next processing batch in advance. After all coordinate sub-signals of the current batch have been extracted, it exits the loop logic and enters the coordinate sub-signal synchronously sent to the timing iteration calculation unit node. The global coordinate sub-signals of each device and the effective working condition data stream of the same batch are synchronously transmitted to the upper-level timing calculation module, providing the core calculation parameters of the workpiece incoming coordinate deviation for the process timing weighted iteration calculation. The entire process relies on three core steps: unified coordinate mapping, partitioned and isolated storage, and batch-ordered cyclic retrieval. It completes the normalization of spatial coordinate references for multiple devices, and the cyclic retrieval mechanism ensures that the processing batch data is processed in an orderly and uncluttered manner. Each process node is layered and data flows in a unidirectional and orderly manner, building a spatial coordinate data transmission channel from the underlying standardized sensor data to the upper-level time-series optimization algorithm.

[0032] Faced with the challenge of proprietary and incompatible protocols among multiple manufacturers' equipment, the system did not adopt the traditional approach of developing interfaces one by one. Instead, it introduced a pre-defined protocol conversion matrix operation model. During system deployment, a complete set of conversion matrices covering all types of CNC machine tool controllers and their auxiliary device controllers in the workshop is pre-loaded. Each set contains independent conversion matrix units adapted to the native protocol of a specific device. When multi-source raw sensor data streams arrive at the conversion node, the system automatically addresses and matches the corresponding conversion matrix unit based on its frame header identifier. The subsequent conversion process mathematically maps the raw values ​​to a standard bit format. Specifically, the standardized output value B_out,i of the i-th sensor signal is equal to the weighted sum of the i-th row of the conversion matrix corresponding to that device type and the original data vector x_raw,j, plus the fixed offset of the device's native protocol, and then multiplied by the reciprocal of the bit scaling factor of that device protocol. The core value of this model lies in unifying the diverse physical quantities into a common numerical space, forming standardized operating condition numerical signals, encapsulating them with a unified fixed message length, and storing them in a dedicated buffer of the secondary channel.

[0033] Although the signal format is standardized, sensor data may still contain noise caused by environmental interference or transient spikes. To address this, the system employs a segmented numerical verification algorithm. This algorithm pre-configures a set of reasonable numerical ranges based on physical laws and device characteristics for each type of sensor value. It extracts all values ​​from the standardized operating condition numerical signal message frame by frame, comparing each value with its corresponding range. Only values ​​falling within this range are considered valid sensor values ​​and retained; any interfering values ​​outside the range are discarded. After filtering, the valid values ​​are reassembled and encapsulated into a valid operating condition data stream message, while the original standardized message is retained intact for future reference. This process ensures that the data relied upon for subsequent complex calculations is clean and reliable. The calculation process for a single time-series iterative operation is as follows:

[0034] in, The execution time of the m-th process is adjusted after the t-th iteration. The spindle loss weighting factor. This represents the real-time spindle wear value for the m-th process. This is the weighting coefficient for coordinate deviation. Let m be the global coordinates of the workpiece. Here, t represents the standard global coordinates of the workpiece, and t represents the current iteration round. M represents the original standard processing time for each process, and M represents the total number of processes in a single batch. The outer summation operator corrects the processing time of all processes in a single batch sequentially. The square operator amplifies the weight of spindle wear on processing time, the logarithmic operator compresses large coordinate deviations to avoid excessive distortion during time correction, the absolute value operator directly extracts the workpiece positioning offset, and the square root operator achieves adaptive convergence based on iteration rounds. The correction amplitude decreases with each iteration to prevent drastic oscillations in timing values. After each iteration, a timing signal with a batch identifier is generated and stored in a three-level transmission channel partitioned cache. Multiple versions of timing signals generated from multiple iterations are stored independently in layers, updating only the data for the current batch to be processed; historical timing messages are not overwritten or deleted. After iterative convergence, the timing signal enters the timing offset compensation stage, where dedicated compensation values ​​are superimposed according to equipment grouping to generate isolated motion execution commands. This formula forms a complete closed-loop data processing link. All the quantitative processing data from the underlying sensor normalization and coordinate unified conversion are transformed into process timings that adapt to real production conditions. If this weighted iterative calculation is canceled, the system can only use fixed standard process durations to arrange the processing timings, and cannot dynamically adapt to equipment wear and workpiece offsets. The timing matching accuracy of multi-machine tool synchronous collaborative processing will drop significantly, and problems such as processing interference, process waiting, and reduced production efficiency are very likely to occur.

[0035] like Figure 4As shown, this flowchart illustrates the full-process control logic of real-time process timing weighted iterative loop and equipment command issuance. The starting node of the process inputs two sets of working condition data: spindle wear and coordinate deviation. This node simultaneously imports two types of core working condition parameters: the spindle wear value is taken from the underlying standardized sensor data stream, and the workpiece coordinate deviation value is taken from the global coordinate difference output by the global coordinate mapping calculation. These two types of parameters correspond to two key production influencing factors: machine tool equipment wear status and workpiece positioning offset. The two sets of data are simultaneously sent to the core calculation node of the real-time process timing weighted iterative timing correction calculation. This node is the core calculation unit of the entire timing optimization process and is pre-configured. Two sets of independent weighted correction coefficients are used to couple the spindle loss value, coordinate deviation value and corresponding weighted coefficients for calculation. Based on multiple composite calculation operators, the original standard execution time of a single process is corrected. After the time of all processes is corrected, the execution order of each machine tool process is rearranged. The corrected process time data of all machine tools is integrated. After the calculation is completed, a single batch update process timing signal is generated and stored in the three-level channel node. The timing signal generated in a single iteration is uniformly bound to a unique batch identifier and stored in the independent cache partition of the three-level transmission channel. The three-level channel is only used to store timing and motion command data and is completely isolated from the underlying sensor data channel and does not interfere with each other. After the timing signal is stored, the loop node for determining the termination of the iteration enters, which controls the start and stop logic of the iteration operation. If the current iteration operation result does not reach the preset convergence threshold, the loop node for recalculating the working condition data is entered. The original working condition data of spindle loss and coordinate deviation are retrieved again to carry out another round of timing iteration operation. Multiple iterations gradually reduce the correction range of the process duration to avoid the process arrangement being disordered due to excessive fluctuations in the correction value in a single correction. When the iteration operation reaches the convergence standard, the loop exits and the iteration is completed. The multi-version timing layered caching node is entered. This node stores the timing signals generated in each iteration layered independently according to the iteration generation order. Each iteration only updates the timing data of the current batch to be processed. The timing data of historical batches is kept in its original storage state and is not overwritten or modified. The timing message of any iteration version can be retrieved according to the batch identifier later.

[0036] After the timing data is stored, it enters the equipment grouping and branching node. It is divided into multiple independent equipment groups according to all CNC machine tools and auxiliary devices in the workshop. The timing data of different groups are processed separately. The timing data of each group enters the group timing superposition dedicated timing offset compensation value node. It matches a dedicated timing offset compensation value set for each group of equipment. It superimposes all the duration values ​​of the timing signal with the compensation values ​​to eliminate the motion response delay and signal transmission delay differences of different equipment. After the superposition operation is completed, it enters the sub-equipment motion execution command sub-signal generation node to generate dedicated command sub-signals for each group of equipment. Each group of sub-signals is independent and there is no cross transmission. After all sub-signals are integrated, they enter the isolated link to send commands to each machine tool and auxiliary device node. Relying on the mutually isolated independent sending links, the dedicated motion commands are transmitted to the corresponding equipment controllers to achieve time alignment, synchronous and collaborative processing of multiple processing equipment. The entire process relies on three layers of logic: iterative cycle optimization of timing, independent compensation for delay in groups, and isolated link instruction issuance. It dynamically adapts to the fluctuations in process duration caused by machine tool wear and workpiece positioning deviation. The data before and after each process node are closely linked, forming a complete closed-loop timing control link from the input of working parameters to the issuance of equipment instructions. This solves the production problems of poor adaptability of fixed timing processing, easy interference in simultaneous processing of multiple devices, and idling waiting.

[0037] To eliminate machining datum deviations caused by differences in local coordinate systems among machine tools, the system constructs a global workpiece coordinate mapping calculation model. This process requires pre-entering a fixed reference origin within the workshop into the system as an absolute reference. During model execution, it simultaneously retrieves the local workpiece coordinate values ​​of a single machine tool and the grasping coordinate values ​​of transfer auxiliary devices (such as robotic arms) from the effective working condition data stream. Then, the model performs a rigorous rigid body transformation on these two sets of coordinates, sequentially performing coordinate translation and rotation conversions to convert the local coordinates based on the equipment's own origin into global coordinates relative to the unified reference origin of the workshop. The mathematical core of this conversion process is calculating a global reference coordinate value G_p, which is based on three-axis standardized sensor coordinate values, combined with the rotation angle of the equipment position relative to the workshop reference origin and the three-axis translation compensation. After all coordinate points of a single piece of equipment have been converted, they are integrated into the global coordinate value set of that equipment. The sets of all related equipment are ultimately summarized to generate a complete, scenario-based global coordinate view signal. This signal is assigned a unique message header identifier and stored in the secondary channel. To avoid data contention, the generated complete global coordinate view signal is split into independent coordinate sub-signals for each device and stored in isolated partitioned cache units. When the process timing calculation unit initiates a retrieval request with a device identifier, the system strictly follows the processing batch order, extracts the sub-signals sequentially from the corresponding partition, and sends them to the calculation unit synchronously with the valid working condition data stream of the same batch.

[0038] While the generated process execution sequence clearly defines the processing order and duration for each device, achieving physical synchronization requires addressing communication delays and response differences between devices. When generating the final sub-device motion execution commands, the system uses a unified timestamp as a benchmark and adds timing offset compensation. Specifically, the system independently assigns a dedicated set of timing offset compensation values ​​to each CNC machine tool and auxiliary device in the workshop. When the process execution sequence signal of a certain device group is retrieved, its duration value is superimposed with the group's compensation value to generate a command sub-signal adapted to the specific device's response characteristics. After all device group sub-signals are integrated to form a complete motion execution command signal, it is directly sent to each device controller through isolated transmission links in a three-level channel, ensuring deterministic command delivery and preventing crosstalk.

[0039] Based on this approach, the constructed system consists of five core modules. The multi-dimensional signal acquisition module is responsible for acquiring and aggregating all sensor signals. The hierarchical data stream transmission module implements strict three-level channel control. The computation and processing module integrates four core units: protocol conversion, numerical verification, coordinate mapping, and timing iteration, serving as the system's intelligent computing hub. The timing instruction generation module is responsible for final synchronization compensation and instruction generation. The distributed timing storage module provides local storage of signals bound to batch timing numbers throughout the process, and can reassemble fragments and restore complete messages based on retrieval requests. These five modules independently execute their respective functions, transmitting only standardized signals conforming to the corresponding hierarchical standards between modules. The transmission links are isolated from each other, collectively forming a highly cohesive and loosely coupled CNC machine tool automation collaborative control system.

[0040] like Figure 5As shown, the present invention also provides an automated collaborative control system for CNC machine tools, including a multi-dimensional signal acquisition module, which acquires various sensor signals from the machine tool and auxiliary devices and outputs multi-source raw sensor data streams; a hierarchical data stream transmission module, which sets up three independent transmission channels to carry raw sensor data streams, standardized operating condition numerical signals, and timing command signals respectively; and a calculation and processing module, which integrates a protocol conversion unit, a numerical verification unit, a coordinate mapping unit, and a timing iteration unit, and receives the output messages from the hierarchical data stream transmission module to sequentially generate standardized operating condition values. The system includes five modules: signal, effective working condition data stream, global coordinate view signal, and process execution timing signal; a timing instruction generation module, which retrieves the process execution timing signal, superimposes the timing offset compensation value to obtain the sub-equipment motion execution instruction signal, and outputs it to the external processing equipment; and a distributed timing storage module, which binds batch timing numbers to various signals to complete local storage of fragments. Upon receiving a retrieval request, it integrates the fragments according to the number to restore the complete signal message. The five modules independently complete the corresponding signal processing operations, and only standardized signals at the matching level are transmitted between modules. The transmission links of each module are isolated from each other.

[0041] The present invention discloses an automated collaborative control method and system for CNC machine tools. Through multi-dimensional sensing and acquisition, heterogeneous signals are normalized using a protocol conversion matrix and segmented verification and noise reduction are performed. A global coordinate mapping is established to unify the coordinates of each machine tool. Based on the weighted iterative sorting of spindle loss and coordinate deviation, the process timing is optimized, and time-aligned sub-equipment motion commands are generated by superimposing timing offset compensation. These commands are then distributed through a three-level isolated transmission channel to achieve synchronous collaborative machining of multiple machines.

[0042] Therefore, the CNC machine tool automated collaborative control method and system of the present invention can solve the problem of the inability to achieve signal normalization and coordinate unification for collaborative control of heterogeneous CNC machine tools.

[0043] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for automated collaborative control of CNC machine tools, characterized in that, include: S1: Collect spindle load signals, feed axis position signals, tool magazine positioning signals, workpiece visual coordinate signals, and motion status signals of auxiliary devices from multiple CNC machine tools through multi-dimensional sensing units to form a multi-source raw sensor data stream; S2: Using the multi-source raw sensor data stream, the heterogeneous device signal normalization conversion is completed through a preset protocol conversion matrix operation model to form a standardized operating condition numerical signal; noise interference signals are removed through a segmented numerical verification algorithm to form an effective operating condition data stream. S3: Using the effective working condition data stream, the conversion from local coordinates of each machine tool to global coordinates is completed through the global workpiece coordinate mapping calculation model, forming a global coordinate view signal; S4: Using the effective working condition data stream and global coordinate view signal, generate the updated process execution timing signal for each machine tool through a real-time process timing weighted iterative sorting algorithm; S5: Using the updated process execution timing signal, add timing offset compensation based on a unified timestamp to form time-aligned sub-equipment motion execution command signals, and output them to each CNC machine tool and auxiliary device to complete synchronous collaborative processing.

2. The method according to claim 1, characterized in that, After acquiring multi-source raw sensor data streams, a unique fixed data frame header identifier is configured for each type of sensor signal. The multi-source raw sensor data streams carrying different header identifiers are divided into three independent data transmission channels to complete layered transmission. The first-level channel only transmits the multi-source raw sensor data streams without conversion processing. The second-level channel only transmits the standardized chemical condition numerical signals after normalization conversion. The third-level channel only stores various timing signals and motion execution command signals to be sent. The three types of data transmission channels are configured with independent transmission buffer spaces that are not interchangeable. Various types of messages can only be transferred within the corresponding level channel. Only one-way data message transfer operations are allowed between different level channels. The corresponding level of operation is only performed after the message header identifier is identified at the cross-level transfer node. Throughout the process, different types of data streams are prevented from being mixed and stored or parsed synchronously in the same transmission buffer space.

3. The method according to claim 2, characterized in that, Before performing the heterogeneous device signal normalization conversion operation, complete conversion matrix groups corresponding to different types of CNC machine tool controllers and their supporting auxiliary device controllers are pre-entered. Each conversion matrix group contains independent conversion matrix units adapted to the native signal protocols of various devices. When retrieving multi-source raw sensor data streams, the conversion matrix unit corresponding to the device type is matched based on the frame header identifier carried by the data stream. All values ​​within a single frame of raw sensor data stream are sequentially substituted into the matched conversion matrix unit to complete the unified bit format conversion. After all sensor values ​​in a single message have undergone matrix conversion, they are uniformly encapsulated to generate a complete standardized operating condition numerical signal message. All message encapsulation operations are performed according to a unified fixed message length standard. All converted standardized operating condition numerical signals are stored in the dedicated buffer space corresponding to the secondary data transmission channel, awaiting subsequent verification and processing. The bit format conversion calculation process is as follows: in, The normalized bit output value of the i-th sensor signal. Let be the bit scaling factor for the protocol of the k-th device. Let the coefficient of the transformation matrix unit of the k-th type of equipment be the matrix coefficient in the i-th row and j-th column. For the j-th channel of the single-frame raw data stream, is the fixed offset of the native protocol for the k-th type of device, and n is the total number of sensor signals contained in a single frame message.

4. The method according to claim 1, characterized in that, Before executing the global workpiece coordinate mapping calculation model, the coordinate value set corresponding to the fixed reference origin in the workshop is pre-entered. The local workpiece coordinate values ​​of a single machine tool and the coordinate values ​​captured by the transfer auxiliary device, contained in the effective working condition data stream, are retrieved. These two sets of coordinate values ​​are synchronously input into the global workpiece coordinate mapping calculation model. The model sequentially performs coordinate translation and rotation conversion operations to complete the conversion of local coordinates to the unified global coordinates of the workshop. After the conversion of all workpiece coordinate values ​​of a single machine, they are integrated into a single machine global coordinate value set. After the global coordinate value sets corresponding to multiple machines are summarized and integrated, a complete global coordinate view signal is generated. The generated global coordinate view signal is encapsulated with a dedicated message header identifier and stored in the secondary data transmission channel cache space for subsequent process timing calculation calls. The calculation formula for the global coordinate value conversion is as follows: in, Let p be the global reference coordinate value of the workpiece on the p-th device. For three-axis standardized sensor coordinate values, This refers to the rotation angle of the equipment station relative to the workshop reference origin. , , The amount of three-axis translation compensation for the workshop reference origin.

5. The method according to claim 4, characterized in that, After generating the global coordinate view signal, the complete global coordinate view signal is split into independent coordinate sub-signals for each device and stored in isolated partitioned buffer units within the secondary data transmission channel. When receiving a signal retrieval request from the process timing calculation unit, the device coordinate sub-signal stored in the corresponding partitioned buffer unit is extracted based on the device identifier carried in the request. The extracted coordinate sub-signals are synchronously transmitted to the process timing weighted iterative sorting algorithm calculation unit along with the valid working condition data stream of the same batch. All coordinate sub-signal retrieval operations can only be executed sequentially according to the processing batch order. The coordinate signal extraction and transmission operation for the next processing batch will not be started until all coordinate sub-signals of the current batch have been retrieved. During the transmission process, the original complete message of the global coordinate view signal is continuously retained in the buffer unit without being overwritten or deleted.

6. The method according to claim 5, characterized in that, When the real-time process timing weighted iterative sorting algorithm is started, the sets of values ​​corresponding to spindle wear and workpiece incoming coordinate deviation are first extracted from the effective working condition data stream. Independent weighted correction coefficient sets are configured for each of these two sets of values. These two sets of values ​​and their corresponding weighted correction coefficient sets are then synchronously input into the real-time process timing weighted iterative sorting algorithm to perform a single timing iteration operation. During this single iteration, the original standard execution time of each process is corrected based on the weighted correction coefficients. After all process time values ​​are corrected, the execution order of the processes corresponding to each machine tool is rearranged. After the order is rearranged, the corrected process time values ​​of all machine tools are integrated to generate complete updated process execution timing signals for each machine tool. The process execution timing signals generated by the single iteration operation are uniformly labeled with batch identifiers and stored in the three-level data transmission channel buffer unit. The calculation process of the single timing iteration operation is as follows: in, The execution time of the m-th process is adjusted after the t-th iteration. This is the weighting factor for spindle loss. This represents the real-time spindle wear value for the m-th process. This is the weighting coefficient for coordinate deviation. Let m be the global coordinates of the workpiece. Here, t represents the standard global coordinates of the workpiece, and t represents the current iteration round. M represents the original standard duration of the process, and M represents the total number of processes in a single batch.

7. The method according to claim 6, characterized in that, After generating the updated process execution timing signals for each machine tool, multiple independent timing signal cache partitions are divided according to the equipment type. The process execution timing signals corresponding to different machine tools are stored in their respective dedicated cache partitions. Each iteration only updates the timing signal value set corresponding to a single batch of processing. The timing signal value sets of other historical batches are kept in their original storage state without being overwritten or modified. When starting the next iteration, only the process execution timing signals corresponding to the current batch to be processed, the corresponding valid working condition data stream, and the global coordinate view signals are read. The multiple versions of process execution timing signals generated by multiple iterations are stored in layers according to the iteration generation order. When retrieving timing signals, the timing signal message generated by any iteration can be extracted based on the iteration batch identifier.

8. The method according to claim 3, characterized in that, When performing segmented numerical verification algorithm processing on standardized operating condition numerical signals, all sensor values ​​within the standardized operating condition numerical signal message are extracted frame by frame. The pre-configured numerical range set corresponding to various sensor values ​​is retrieved. Each sensor value in a single frame message is compared with the corresponding numerical range set in turn. After comparison, sensor values ​​that are within the corresponding range are retained, while interference sensor values ​​that exceed the range are removed. All retained valid sensor values ​​are re-integrated and encapsulated to form a complete single-frame valid operating condition data stream message. All messages that have completed the verification operation are uniformly summarized and integrated to generate a complete valid operating condition data stream overall message and transmitted to the secondary data transmission channel to await coordinate conversion calculation. During the verification process, the original standardized operating condition numerical signal message is completely retained without deletion.

9. The method according to claim 8, characterized in that, After generating the updated process execution timing signal, the equipment is divided into independent groups according to all CNC machine tools and auxiliary devices in the workshop. Each group of equipment is assigned a dedicated set of timing offset compensation values. After retrieving the process execution timing signal of the corresponding equipment group, the entire duration value of the timing signal is superimposed with the dedicated timing offset compensation value set of the group. After the superposition operation is completed, a sub-equipment motion execution command sub-signal is generated to adapt to the single group of equipment. After the motion execution command sub-signals corresponding to all equipment groups are integrated to form a complete motion execution command signal, they are sent to the corresponding equipment controllers according to the independent buffer partitions of the three-level data transmission channel. The command sub-signal sending links of different equipment groups are isolated from each other and there is no message cross-transmission operation process.

10. A system using the automated collaborative control method for CNC machine tools according to any one of claims 1-9, characterized in that, include: The multi-dimensional signal acquisition module collects various sensor signals from machine tools and auxiliary devices and outputs multi-source raw sensor data streams. The layered data stream transmission module is equipped with three independent transmission channels that respectively carry the raw sensor data stream, standardized operating condition numerical signals, and timing command signals. The computation and processing module integrates a protocol conversion unit, a numerical verification unit, a coordinate mapping unit, and a timing iteration unit. It receives the output messages from the layered data stream transmission module and sequentially generates standardized working condition numerical signals, effective working condition data streams, global coordinate view signals, and process execution timing signals. The timing instruction generation module retrieves the process execution timing signal, superimposes the timing offset compensation value to obtain the sub-equipment motion execution instruction signal, and outputs it to the external processing equipment. The distributed time-series storage module binds batch time sequence numbers to various signals to complete local storage of fragments. Upon receiving a retrieval request, it integrates the fragments according to the number to restore the complete signal message. The five modules independently complete the corresponding signal processing operations. Only standardized signals at the matching level are transmitted between modules, and the transmission links of each module are isolated from each other.

Images