Smart sensor driven watch timekeeping accuracy calibration system

Abstract

This invention provides an intelligent sensor-driven clock timing accuracy calibration system, relating to the field of data processing technology. It acquires raw temperature sequence data with time stamps and location information by deploying temperature sensing units at different heating locations on the clock timing device; identifies the dominant heating region based on the temperature change rate and duration and generates corresponding weights; analyzes the heat transfer time difference in conjunction with the crystal oscillator installation position to construct a local thermal effect response sequence with time lag; derives the crystal oscillator frequency evolution process based on this response sequence and generates a predicted frequency offset; pre-adjusts the timing pulse before errors occur to form a timing reference signal; and simultaneously corrects the dominant heat source weights and heat transfer delay relationship by comparing with an external standard time source, achieving dynamic calibration of timing accuracy.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a smart sensor-driven clock timing accuracy calibration system. Background Technology

[0002] The closest existing technologies to this patent mostly employ electronic clock timing circuits in conjunction with a reference oscillation source (such as a quartz crystal) to achieve a time reference. They are calibrated periodically with an external standard time source (such as GPS or radio timing signals) or compensated for the oscillation frequency through software algorithms. In addition, some solutions introduce temperature-compensated crystal oscillators (TCXOs) or simple temperature sensors to correct frequency deviations based on changes in ambient temperature, thereby improving timing accuracy.

[0003] However, in applications such as large outdoor public clocks (e.g., station tower clocks), this technology has significant drawbacks: when the device is in a complex environment (large temperature differences between day and night, uneven sunlight), relying solely on single temperature compensation or intermittent timekeeping cannot reflect changes in local temperature gradients in real time. For example, if one side of the clock is exposed to direct sunlight while the other is in shadow, the actual operating temperature of the crystal oscillator may differ from the measured temperature, resulting in cumulative errors that could lead to deviations of several seconds over several hours, affecting the accuracy of the timekeeping. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent sensor-driven clock timing accuracy calibration system, which aims to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] A smart sensor-driven watch timing accuracy calibration system, the system comprising:

[0007] The spatial orientation acquisition module is used to deploy temperature sensing units in different heated orientations of the clock timing device, collect the spatial position of each temperature sensing unit and the corresponding real-time temperature data, and generate raw temperature sequence data with time stamps and orientation identifiers.

[0008] The heat source dominance identification module is used to calculate the rate and duration of temperature change in each direction based on the original temperature sequence data, identify the dominant heated area that has the main impact on the crystal oscillator, and generate the corresponding dominant heat source identifier and its influence weight.

[0009] The heat transfer delay modeling module is used to analyze the temperature change transfer time difference between the dominant heat source and the crystal oscillator position based on the dominant heat source identification and the crystal oscillator installation position, and to construct a local thermal effect response sequence that includes time lag relationships.

[0010] The predicted frequency offset generation module is used to deduce the frequency evolution sequence of the crystal oscillator in the subsequent time period based on the local thermal effect response sequence, and generate the corresponding predicted frequency offset.

[0011] The feedforward calibration control module is used to pre-adjust the cumulative counting rhythm of the timing pulses before the timing error actually occurs, based on the predicted frequency offset, and generate a feedforward corrected timing reference signal.

[0012] The closed-loop adaptive correction module is used to compare the timing reference signal with an external standard time source and correct the dominant heat source weight and heat transfer delay relationship to continuously optimize prediction accuracy.

[0013] Preferably, the heat source dominance identification module includes:

[0014] The multi-source competitive evaluation unit is used to jointly analyze the magnitude, rate, and duration of temperature changes in each location within a continuous time period in the original temperature sequence data, forming a competitive relationship sequence of the influence of temperature changes in each location on the crystal oscillator temperature.

[0015] The dominant switching determination unit is used to identify the dominant heated area in the current time period in the competition relationship sequence, and generate a dominant heat source switching identifier when the degree of influence between different directions alternates, and record the corresponding switching time period.

[0016] The dynamic weight allocation unit is used to assign a time-varying weight to the corresponding dominant heat source identifier based on the continuity and superposition of the influence of each direction before and after the switch during the process of switching the dominant heat source.

[0017] Preferably, the heat transfer delay modeling module includes:

[0018] The multi-path partitioning unit is used to divide the heat transfer process from the dominant heated area to the crystal oscillator position into multiple paths with different structures based on the spatial structure between the dominant heat source identifier and the crystal oscillator installation position.

[0019] The path difference analysis unit is used to analyze the start time and completion time of temperature change transmission in each path, and determine the independent transmission delay interval for each path.

[0020] The delay competition superposition unit is used to superimpose the thermal effects of different paths in time according to the propagation delay interval of each path and the order in which they arrive at the crystal oscillator position, forming a local thermal effect response sequence that includes the sequential action relationship of multiple paths.

[0021] Preferably, the predicted frequency offset generation module includes:

[0022] The delay chain expansion unit is used to expand the crystal oscillator temperature change process in stages according to the arrival order of thermal effects along different paths in the local thermal effect response sequence, forming a temperature effect evolution sequence containing multiple time periods.

[0023] The stage superposition derivation unit is used to perform stage-by-stage superposition derivation of the crystal oscillator frequency change process according to the sequential relationship of each stage in the temperature influence evolution sequence, forming a frequency evolution sequence that reflects the frequency change with time in the future time period.

[0024] The forward offset determination unit is used to determine the predicted frequency offset for the corresponding time period based on the frequency change portion of the corresponding future time period in the frequency evolution sequence.

[0025] Preferably, the feedforward calibration control module includes:

[0026] The offset distribution generation unit is used to segment and expand the frequency changes in future time periods according to the predicted frequency offset and in chronological order, so as to form the frequency offset distribution sequence corresponding to each time period.

[0027] The rhythm mapping generation unit is used to map the frequency offset of each time period to the adjustment rhythm of the timing pulse cumulative count according to the frequency offset distribution sequence, and generate the corresponding counting rhythm control sequence.

[0028] The pre-adjustment execution unit is used to pre-adjust the cumulative counting process of timing pulses before the corresponding time period arrives according to the counting rhythm control sequence, and generate an initial timing reference signal corresponding to the frequency offset distribution sequence.

[0029] The continuous constraint correction unit is used to perform continuous constraint processing on the initial timing reference signal according to the rhythm change relationship between adjacent time periods in the counting rhythm control sequence, correct the transition of counting changes between adjacent time periods, and generate a continuously changing timing reference signal.

[0030] Preferably, the multi-source competition evaluation unit includes:

[0031] The feature set generation subunit is used to correlate and combine the magnitude, rate and duration of temperature change in each location in the original temperature sequence data over a continuous time period to generate a temperature change feature set corresponding to each location.

[0032] The priority sequence generation subunit is used to sort the influence of temperature changes on crystal oscillator temperature in each direction according to the order and intensity of changes in each direction within the same time period in the temperature change feature set, and generate a competitive relationship sequence with priority order.

[0033] The competition state generation subunit is used to determine whether the competition relationship is in a stable or changing state based on the change of priority of each position in the competition relationship sequence over time, generate the corresponding competition state identifier, and record the continuous influence state of the dominant heated area in the stable state in the subsequent time period.

[0034] Preferably, the dominant switching determination unit includes:

[0035] The dominant region generation sub-unit is used to determine the dominant heating region in the current time period based on the competition status identifier and the highest priority orientation in the competition relationship sequence, and to generate the dominant region identifier.

[0036] The switching process generation subunit is used to identify the alternation process between the original dominant heating area and the new dominant heating area based on the change of the dominant area identifier between adjacent time periods, and to generate a dominant switching process sequence.

[0037] The switching identifier generation subunit is used to determine the corresponding time period when the original dominant heating area switches to the new dominant heating area based on the dominant switching process sequence, and to generate a dominant heat source switching identifier. At the same time, it records the residual impact information of the original dominant heating area after the switch, so as to be called in subsequent weight allocation.

[0038] Preferably, the dynamic weight allocation unit includes:

[0039] The stage weight generation subunit is used to calculate the weights of the original dominant heating area, the new dominant heating area, and the residual influence area based on the dominant switching process sequence, the dominant heat source switching identifier, and the residual influence record data, and generate the corresponding stage effect weight sequence.

[0040] The continuous weight generation subunit is used to continuously adjust the corresponding weights of each direction according to the time-varying relationship of the phased weight sequence, so that the weight changes during the switching of the dominant heat source remain continuous and generate a weight distribution sequence that changes continuously over time.

[0041] The coupling weight generation subunit is used to couple the weights of each direction based on the mutual influence relationship between different directions in the weight distribution sequence and in combination with the dominant heat source switching identifier. This couples the weights of each direction to form a mutual constraint relationship between the original dominant heating area, the new dominant heating area, and the residual influence area, generating a composite weight sequence that reflects the combined effect of multiple directions. This composite weight sequence is then used as the weight corresponding to the dominant heat source identifier.

[0042] Preferably, the path difference parsing unit includes:

[0043] The path time series generation subunit is used to extract the start and end times of temperature changes in each path based on the temperature change process in each path over a continuous time period, and generate the corresponding path time series data.

[0044] The delay interval generation sub-unit is used to perform interval calculation processing on the temperature transfer process of each path based on the time difference between the start and end times of the temperature change of each path in the path time series data, and generate the corresponding path delay interval.

[0045] The state association correction subunit is used to perform state association correction processing on the path delay interval according to the temperature change amplitude and trend during the temperature change process of each path, so that the delay interval of each path is adjusted with the temperature change state, and the independent transmission delay interval corresponding to each path is determined.

[0046] Preferably, the delay contention stacking unit includes:

[0047] The arrival order generation sub-unit is used to sort the time order of the thermal effects of different paths to the crystal oscillator position according to the independent transmission delay interval, and generate the path arrival order sequence.

[0048] The state transfer generation sub-unit is used to generate a state transfer sequence between paths by taking the temperature state formed at the crystal oscillator position of the first arriving path as the initial state of the thermal influence of the later arriving path, according to the path arrival sequence.

[0049] The nonlinear superposition generation sub-unit is used to successively superimpose the thermal effects of different paths on the basis of the previous state according to the state transfer sequence between paths, so that the effect of the later arriving path depends on the temperature state formed by the earlier arriving path, and generates a local thermal effect response sequence containing the interdependence between paths.

[0050] The above-described solution of the present invention has at least the following beneficial effects:

[0051] By deploying temperature sensing units at different heating locations in the clock timing device, raw temperature sequence data with time and spatial distribution characteristics can be obtained, enabling the change of ambient temperature to be expressed in a multi-directional form. Compared with the single-point temperature acquisition method, it can reflect the actual situation of uneven local heating, thereby avoiding the inconsistency between the measured temperature and the actual heating state of the crystal oscillator.

[0052] Based on this, by identifying the dominant heating area and assigning corresponding weights, the effects of different heat sources on the crystal oscillator can be differentiated and processed, enabling timing adjustments to respond to the main sources of influence, thereby reducing interference from non-dominant temperature changes.

[0053] Furthermore, by combining the analysis of heat transfer time difference with the crystal oscillator installation position, a local thermal effect response sequence with time lag is constructed, so that the crystal oscillator heating process can reflect the temperature change transfer path and delay characteristics, thereby making the correction of the timing reference no longer dependent on the instantaneous ambient temperature.

[0054] Based on this, by deriving the crystal oscillator frequency evolution process and generating the predicted frequency offset, the timing system can obtain the frequency change trend before the error occurs and pre-adjust the timing pulse, thereby reducing the error accumulation process.

[0055] Meanwhile, by comparing with an external standard time source and correcting the dominant heat source weight and heat transfer delay relationship, the system can continuously update parameters during long-term operation, ensuring that timing accuracy remains stable in complex environments. Attached Figure Description

[0056] Figure 1 This is an architecture diagram of an intelligent sensor-driven clock timing accuracy calibration system provided in an embodiment of the present invention. Detailed Implementation

[0057] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0058] like Figure 1 As shown, an embodiment of the present invention proposes an intelligent sensor-driven clock timing accuracy calibration system, the system comprising:

[0059] The spatial orientation acquisition module is used to deploy temperature sensing units in different heated orientations of the clock timing device, collect the spatial position of each temperature sensing unit and the corresponding real-time temperature data, and generate raw temperature sequence data with time stamps and orientation identifiers.

[0060] The heat source dominance identification module is used to calculate the rate and duration of temperature change in each direction based on the original temperature sequence data, identify the dominant heated area that has the main impact on the crystal oscillator, and generate the corresponding dominant heat source identifier and its influence weight.

[0061] The heat transfer delay modeling module is used to analyze the temperature change transfer time difference between the dominant heat source and the crystal oscillator position based on the dominant heat source identification and the crystal oscillator installation position, and to construct a local thermal effect response sequence that includes time lag relationships.

[0062] The predicted frequency offset generation module is used to deduce the frequency evolution sequence of the crystal oscillator in the subsequent time period based on the local thermal effect response sequence, and generate the corresponding predicted frequency offset.

[0063] The feedforward calibration control module is used to pre-adjust the cumulative counting rhythm of the timing pulses before the timing error actually occurs, based on the predicted frequency offset, and generate a feedforward corrected timing reference signal.

[0064] The closed-loop adaptive correction module is used to compare the timing reference signal with an external standard time source and correct the dominant heat source weight and heat transfer delay relationship to continuously optimize prediction accuracy.

[0065] In this embodiment of the invention, by deploying temperature sensing units at different heated locations on the clock timing device, raw temperature sequence data with time stamps and spatial location information is obtained. This allows the change in ambient temperature to be expressed in a spatially distributed form rather than as a single point, thereby reflecting the temperature differences between different heated areas and their changes over time, providing basic data support for subsequent analysis of the actual heated state of the crystal oscillator.

[0066] Based on the original temperature sequence data, a comprehensive analysis of the rate and duration of temperature change in each direction is performed to identify the heated areas that have the main impact on the crystal oscillator and generate corresponding dominant heat source identifiers and their weights. This allows the impact of different heat sources in the environment on the crystal oscillator to be distinguished and quantified, avoiding the simple equivalent treatment of temperature changes in each direction. As a result, the adjustment of the timing reference can respond to the main sources of influence.

[0067] After determining the dominant heated area, and considering the installation position of the crystal oscillator in the clock timing device, the time difference of heat transfer from the heated area to the crystal oscillator position is analyzed. A local thermal effect response sequence containing time lag is constructed, so that the actual temperature change of the crystal oscillator no longer depends solely on the current ambient temperature, but can reflect the temperature change transmission process and its time delay characteristics, thereby improving the accuracy of the description of the crystal oscillator's working state.

[0068] Based on this, the frequency change process of the crystal oscillator in the subsequent time period is deduced according to the local thermal effect response sequence, forming a corresponding frequency evolution sequence, and further generating a predicted frequency offset, so that the timing system can obtain the future frequency change trend before the error occurs, thereby providing a basis for adjusting the timing reference in advance.

[0069] Based on the predicted frequency offset, the cumulative counting rhythm of the timing pulses is pre-adjusted to generate a feedforward corrected timing reference signal, so that the timing error is canceled before it is formed, avoiding the lag problem caused by relying solely on compensation methods after the error occurs, thus enabling the timing process to be synchronously adjusted with changes in the environment.

[0070] Meanwhile, by comparing the corrected timing reference signal with an external standard time source, the weight of the dominant heat source and the relationship between heat transfer delay are continuously corrected, enabling the system to continuously correct its own parameters during long-term operation and ensuring consistency in the timing adjustment process under different environmental conditions.

[0071] The following explanation is based on specific scenarios:

[0072] Taking a large outdoor public clock as an example, one side is exposed to sunlight while the other side is in shadow. The temperature sensing unit acquires temperature change information from different directions, analyzes it to determine that the side exposed to sunlight is the dominant heat-affected area, and calculates the time delay of heat transfer to the crystal oscillator based on the crystal oscillator's installation position, forming a corresponding thermal effect response sequence. Then, based on this sequence, the frequency change of the crystal oscillator in the subsequent time period is deduced, and the timing pulse is adjusted before timing errors occur. At the same time, during operation, relevant parameters are continuously corrected by comparing with the standard time, so that the device can maintain stable timing results even in an environment with alternating sunlight and shadow.

[0073] In a preferred embodiment of the present invention, the spatial orientation acquisition module includes:

[0074] The orientation division process involves dividing the outer surface and internal space of a clockwork timing device into multiple heated areas with distinct directional attributes, based on the device's external structure, and assigning a corresponding orientation marker to each heated area.

[0075] The sensor unit deployment steps are as follows: install temperature sensor units in each heated area and record the spatial coordinate information of each temperature sensor unit relative to the center position of the clock timing device.

[0076] The data acquisition process involves acquiring the temperature values ​​of each temperature sensing unit within a continuous time period according to a preset sampling cycle, and simultaneously recording the acquisition time to form a time stamp.

[0077] The data association step associates each temperature value with its corresponding directional marker and spatial coordinates to form a temperature data set with spatial and temporal attributes.

[0078] The sequence generation step involves arranging the temperature data set within a continuous time period in chronological order to generate raw temperature sequence data with time stamps and location identifiers.

[0079] In a preferred embodiment of the present invention, the heat source dominance identification module includes:

[0080] The rate of change calculation step involves calculating the ratio of the temperature change in each location to the corresponding time interval between adjacent time sampling points to obtain the degree of temperature change in each location.

[0081] The duration determination step involves statistically analyzing the time periods during which the temperature in each direction is in a state of continuous rise or continuous fall, to obtain the duration of temperature changes in each direction.

[0082] The impact assessment step combines the rate of temperature change with the duration of the change, assigning a higher impact level to locations with a faster rate of change and a longer duration of change.

[0083] The process for determining the dominant heating area involves comparing the degree of influence in each direction within the same time period and selecting the direction with the highest degree of influence as the dominant heating area.

[0084] The weighting process involves assigning corresponding weights to each area based on its proportion of influence in the overall picture, resulting in higher weights for the dominant heated areas and relatively lower weights for other areas.

[0085] In a preferred embodiment of the present invention, the closed-loop adaptive correction module is used to compare the timing reference signal with an external standard time source, correct the dominant heat source weight and heat transfer delay relationship, so as to continuously optimize the prediction accuracy, including:

[0086] The benchmark comparison step involves comparing the timing result corresponding to the timing benchmark signal with the standard time provided by an external standard time source every moment during continuous operation to obtain the timing deviation within each time period.

[0087] The deviation extraction step extracts the changes in timing deviation over a continuous time period, identifying the direction, magnitude, and duration of the timing deviation changes.

[0088] The source correlation step involves correlating the changes in timing deviation with the dominant heat source identifier, influence weight, and path delay interval within the corresponding time period to determine the main sources of influence causing timing deviation.

[0089] The weighting correction step adjusts the weight of the corresponding dominant heating area to match the actual degree of influence when the timing deviation continues to increase.

[0090] The delay correction step adjusts the transfer delay interval of the corresponding path when the trend of timing deviation does not match the heat transfer time, so that the delay relationship is consistent with the actual heat transfer process.

[0091] The parameter update step updates the corrected action weights and transmission delay intervals to the heat source dominance identification module and the heat transmission delay modeling module, so that the subsequent prediction process is calculated based on the updated parameters.

[0092] The closed-loop output step continues to execute the above steps during subsequent operation, enabling the generation process of the timing reference signal to be adaptively corrected as the environment changes.

[0093] In a preferred embodiment of the present invention, the heat source dominance identification module includes:

[0094] The multi-source competitive evaluation unit is used to jointly analyze the magnitude, rate, and duration of temperature changes in each location within a continuous time period in the original temperature sequence data, forming a competitive relationship sequence of the influence of temperature changes in each location on the crystal oscillator temperature.

[0095] The dominant switching determination unit is used to identify the dominant heated area in the current time period in the competition relationship sequence, and generate a dominant heat source switching identifier when the degree of influence between different directions alternates, and record the corresponding switching time period.

[0096] The dynamic weight allocation unit is used to assign a time-varying weight to the corresponding dominant heat source identifier based on the continuity and superposition of the influence of each direction before and after the switch during the process of switching the dominant heat source.

[0097] In this embodiment of the invention, the multi-source competition evaluation unit jointly analyzes the amplitude, rate, and duration of temperature changes in different directions to form a competitive relationship sequence of the influence of each direction on the crystal oscillator temperature. This allows the interaction between heat sources to be distinguished according to time and intensity, thereby avoiding the simple superposition of the influence of different heat sources. The dominant switching determination unit identifies the change process of the dominant heating region based on the competitive relationship sequence and generates a dominant heat source switching identifier, so that the dominant relationship of heat sources is clearly expressed in the time dimension. The dynamic weight allocation unit continuously adjusts the weight of each direction according to the switching process, so that different heat sources maintain a continuous transition during the alternation process. This allows the description of the crystal oscillator's heating state to be dynamically updated with the change of dominant position, avoiding abrupt changes in timing adjustment caused by the switching of dominant position.

[0098] In a preferred embodiment of the present invention, the heat transfer delay modeling module includes:

[0099] The multi-path partitioning unit is used to divide the heat transfer process from the dominant heated area to the crystal oscillator position into multiple paths with different structures based on the spatial structure between the dominant heat source identifier and the crystal oscillator installation position.

[0100] The path difference analysis unit is used to analyze the start time and completion time of temperature change transmission in each path, and determine the independent transmission delay interval for each path.

[0101] The delay competition superposition unit is used to superimpose the thermal effects of different paths in time according to the propagation delay interval of each path and the order in which they arrive at the crystal oscillator position, forming a local thermal effect response sequence that includes the sequential action relationship of multiple paths.

[0102] In this embodiment of the invention, the multi-path partitioning unit divides the heat transfer process from the dominant heated area to the crystal oscillator position into multiple paths, so that heat transfer is no longer described by a single path; the path difference analysis unit determines the independent transfer delay interval corresponding to each path by analyzing the start and end times of temperature changes in each path, so that the transfer characteristics of different paths can be distinguished; the delay competition superposition unit superimposes the thermal effects according to the delay intervals of each path and the arrival order, so that the thermal effects of different paths participate in the calculation in time order, thereby reflecting the influence relationship of the sequential effects in the multi-path transfer process and avoiding simplifying multi-path transfer to synchronous effects.

[0103] In a preferred embodiment of the present invention, the multi-path partitioning unit includes:

[0104] The path identification step identifies the possible heat transfer channels based on the structural connection between the dominant heated area and the crystal oscillator mounting location, including paths conducted through the outer casing, paths conducted through air convection, and paths conducted through the internal support structure.

[0105] The path segmentation step involves segmenting each transmission channel according to material changes or structural nodes, so that each segment has relatively consistent transmission characteristics.

[0106] The path identification step involves setting corresponding identification information for each path and its segments, and recording their start and end positions.

[0107] The path characteristics recording step records the material type, length, and spatial location of each path segment to reflect the transmission differences of different paths;

[0108] The path set generation step summarizes all identified paths and their segmentation information to form a multi-path transmission structure description from the dominant heated area to the crystal oscillator installation location.

[0109] In a preferred embodiment of the present invention, the predicted frequency shift generation module includes:

[0110] The delay chain expansion unit is used to expand the crystal oscillator temperature change process in stages according to the arrival order of thermal effects along different paths in the local thermal effect response sequence, forming a temperature effect evolution sequence containing multiple time periods.

[0111] The stage superposition derivation unit is used to perform stage-by-stage superposition derivation of the crystal oscillator frequency change process according to the sequential relationship of each stage in the temperature influence evolution sequence, forming a frequency evolution sequence that reflects the frequency change with time in the future time period.

[0112] The forward offset determination unit is used to determine the predicted frequency offset for the corresponding time period based on the frequency change portion of the corresponding future time period in the frequency evolution sequence.

[0113] In this embodiment of the invention, the delay chain unpacking unit unpacks the crystal oscillator temperature change process in stages according to the arrival order of thermal effects along different paths, so that the temperature change can form a staged process according to the time sequence; the stage superposition derivation unit derives the crystal oscillator frequency change process stage by stage according to the sequential relationship between each stage, so that the frequency change can correspond to the temperature change transmission process; the advance offset determination unit generates a predicted frequency offset based on the change portion of the future time period in the frequency evolution sequence, so that the timing system can obtain future frequency change information, thereby changing the timing adjustment from a passive response to an advance adjustment based on the change process.

[0114] In a preferred embodiment of the present invention, the delay chain unwinding unit includes:

[0115] The arrival sequence reading step reads the information of the order in which the thermal effects of each path arrive at the crystal oscillator position in the local thermal effect response sequence, and forms the path arrival order according to the time sequence.

[0116] The step of determining the stage start point is to take the moment when the thermal impact of the first path reaches the crystal oscillator position as the first stage start point, and take the moment when the thermal impact of each subsequent path reaches the crystal oscillator position as the next stage start point.

[0117] The step of determining the stage endpoint is to take the end of the time period before the arrival of the adjacent subsequent path thermal effect as the current stage endpoint, and the end of the time period after the completion of the thermal effect of all paths as the final stage endpoint.

[0118] The stage-based impact extraction step extracts the thermal impact content of the path that has reached the crystal oscillator position in each stage, and excludes the thermal impact of the path that has not yet reached the crystal oscillator position from the current stage.

[0119] The stage sequence generation step arranges the thermal effects of the reached paths corresponding to each stage in chronological order to form a temperature effect evolution sequence containing multiple time periods.

[0120] In a preferred embodiment of the present invention, the stage superposition derivation unit includes:

[0121] The temperature reading steps involve sequentially reading the temperature influence status corresponding to each stage according to the temperature influence evolution sequence.

[0122] The frequency change conversion step converts the temperature-affected state at each stage into the corresponding frequency change value based on the frequency change correspondence of the crystal oscillator under different temperature conditions.

[0123] The steps for connecting the preceding and following stages are to use the frequency change results of the current stage as the starting state for the frequency change derivation of the next stage, so that the frequency changes of the subsequent stages are based on the results of the previous stage.

[0124] The derivation steps of stage superposition are to connect the frequency change results corresponding to each stage in chronological order, and to sequentially synthesize the change process between consecutive stages to form a frequency change process covering future time periods.

[0125] The evolution sequence generation step involves associating and arranging the frequency change results of each stage with their corresponding time periods to generate a frequency evolution sequence that reflects the frequency changes over time in the future time period.

[0126] In a preferred embodiment of the present invention, the forward offset determination unit includes:

[0127] The future time period selection step involves selecting subsequent time periods from the frequency evolution sequence that have not yet entered the current timing control moment as the prediction interval;

[0128] The change extraction step extracts the frequency change content corresponding to each time period within the prediction interval, and distinguishes the start and end states of change for each time period.

[0129] The offset determination step involves determining the predicted frequency offset for the corresponding time period based on the degree of deviation of the frequency change results within each time period from the standard timing frequency.

[0130] The offset association step establishes a one-to-one correspondence between each predicted frequency offset and the corresponding future time period, forming offset data arranged in chronological order.

[0131] The result output step outputs the predicted frequency offset for the corresponding future time period, which is used to pre-adjust the cumulative counting rhythm of the timing pulses.

[0132] In a preferred embodiment of the present invention, the feedforward calibration control module includes:

[0133] The offset distribution generation unit is used to segment and expand the frequency changes in future time periods according to the predicted frequency offset and in chronological order, so as to form the frequency offset distribution sequence corresponding to each time period.

[0134] The rhythm mapping generation unit is used to map the frequency offset of each time period to the adjustment rhythm of the timing pulse cumulative count according to the frequency offset distribution sequence, and generate the corresponding counting rhythm control sequence.

[0135] The pre-adjustment execution unit is used to pre-adjust the cumulative counting process of timing pulses before the corresponding time period arrives according to the counting rhythm control sequence, and generate an initial timing reference signal corresponding to the frequency offset distribution sequence.

[0136] The continuous constraint correction unit is used to perform continuous constraint processing on the initial timing reference signal according to the rhythm change relationship between adjacent time periods in the counting rhythm control sequence, correct the transition of counting changes between adjacent time periods, and generate a continuously changing timing reference signal.

[0137] In this embodiment of the invention, the offset distribution generation unit segments the frequency changes within a future time period according to the predicted frequency offset, so that the frequency changes form a corresponding distribution sequence in the time dimension; the rhythm mapping generation unit converts the distribution sequence into the adjustment rhythm of the timing pulse accumulation count, so that the frequency changes can be mapped to a specific timing control mode; the pre-adjustment execution unit pre-adjusts the timing pulses before the corresponding time period arrives, so that the timing reference signal can respond to the frequency change process in advance; the continuous constraint correction unit performs continuous constraint processing on the count changes between adjacent time periods, so that the timing reference signal maintains a smooth transition during the adjustment process, thereby maintaining the continuity of the timing process under frequency changes.

[0138] In a preferred embodiment of the present invention, the offset distribution generation unit includes:

[0139] The prediction data reading steps involve reading the prediction frequency offsets corresponding to each future time period output by the pre-offset determination unit and sorting them according to time sequence.

[0140] The time segmentation step divides the future time period into multiple continuous segments according to a preset control cycle, so that each segment corresponds to a set of frequency offset data to be executed.

[0141] The offset mapping step maps each predicted frequency offset to its corresponding time segment, forming an offset data set for each time segment.

[0142] The distribution and sorting step involves merging the offset data within the same time interval to determine the frequency offset range and offset change direction corresponding to that time interval.

[0143] The sequence generation step involves arranging the frequency offset results corresponding to each time segment in chronological order to generate a frequency offset distribution sequence for each time segment.

[0144] In a preferred embodiment of the present invention, the rhythm mapping generation unit includes:

[0145] The distribution sequence reading steps involve reading the frequency offset distribution sequence corresponding to each time segment and obtaining the offset magnitude and offset direction within each segment.

[0146] The offset conversion step determines, based on the degree to which the frequency deviates from the standard timing frequency within each time segment, the cumulative count that the timing pulse should increase, decrease, or remain unchanged within that time segment.

[0147] The rhythm determination step involves assigning the cumulative count states that increase, decrease, or remain unchanged to multiple control moments within the corresponding time interval, and determining the count adjustment rhythm for each control moment.

[0148] The sequence association step establishes a correspondence between the counting adjustment rhythm of each control moment and its corresponding time segment, forming control data that covers future time periods;

[0149] The control sequence generation step arranges the counting adjustment rhythms within each time segment in chronological order to generate the corresponding counting rhythm control sequence.

[0150] In a preferred embodiment of the present invention, the pre-adjustment execution unit is used to pre-adjust the cumulative counting process of timing pulses before the arrival of the corresponding time period according to the counting rhythm control sequence, and to generate an initial timing reference signal corresponding to the frequency offset distribution sequence, including:

[0151] The control sequence reading step reads the counting adjustment rhythm and corresponding time period information for each future time segment in the counting rhythm control sequence.

[0152] The offset start identification step determines the moment when the frequency offset begins to occur in each time interval based on the start time of the change corresponding to the predicted frequency offset.

[0153] The advance determination step involves determining the advance execution time of the corresponding counting adjustment rhythm based on the time length required for the timing pulse to be adjusted to a stable output and the start time of the frequency offset change, so that the advance execution time is earlier than the start time of the corresponding time segment.

[0154] In the pre-adjustment execution step, at the pre-execution time, a control operation is performed on the cumulative counting process of the timing pulse to increase the count, decrease the count, or maintain the count, so that the timing reference is already in the target adjustment state when the corresponding time segment is reached;

[0155] The reference generation step involves continuously outputting the pre-adjusted timing pulse to the timing circuit to form the initial timing reference signal for the corresponding time segment.

[0156] The correspondence establishment step involves associating each initial timing reference signal with the frequency offset distribution sequence of the corresponding time segment to form initial reference data corresponding in time sequence.

[0157] In a preferred embodiment of the present invention, the continuous constraint correction unit includes:

[0158] The adjacent segment extraction step involves extracting two adjacent time segments from the counting rhythm control sequence and reading the corresponding counting adjustment rhythm.

[0159] The rhythm difference identification step compares the changes in the counting adjustment rhythm between adjacent time segments to identify the location and magnitude of changes in cumulative counting frequency.

[0160] The transition interval setting step involves setting a transition adjustment interval at the connection point between adjacent time intervals, so that the counting adjustment rhythm of the previous time interval gradually changes to the counting adjustment rhythm of the next time interval.

[0161] The continuous correction step gradually corrects the counting change process in the initial timing reference signal within the transition adjustment interval, avoiding abrupt switching between adjacent time segments;

[0162] The reference output step connects the timing reference results of each time segment after the continuity constraint processing in chronological order to generate a continuously changing timing reference signal.

[0163] In a preferred embodiment of the present invention, the multi-source competition evaluation unit includes:

[0164] The feature set generation subunit is used to correlate and combine the magnitude, rate and duration of temperature change in each location in the original temperature sequence data over a continuous time period to generate a temperature change feature set corresponding to each location.

[0165] The priority sequence generation subunit is used to sort the influence of temperature changes on crystal oscillator temperature in each direction according to the order and intensity of changes in each direction within the same time period in the temperature change feature set, and generate a competitive relationship sequence with priority order.

[0166] The competition state generation subunit is used to determine whether the competition relationship is in a stable or changing state based on the change of priority of each position in the competition relationship sequence over time, generate the corresponding competition state identifier, and record the continuous influence state of the dominant heated area in the stable state in the subsequent time period.

[0167] In this embodiment of the invention, the feature set generation subunit combines temperature changes in various directions to form a set that reflects the characteristics of temperature changes, so that temperature changes in different directions have a unified expression form; the priority sequence generation subunit sorts the influence of each direction according to the order and intensity of the changes, so that the influence of each direction on the crystal oscillator can be distinguished according to priority; the competition state generation subunit determines the competition state according to the priority change and records the continuous influence of the dominant heating area in the stable state, so that the heat source effect not only reflects the current state, but also reflects its continued influence in the subsequent time period, thereby providing a basis for the determination of the dominant switching.

[0168] In a preferred embodiment of the present invention, the feature set generation subunit includes:

[0169] The data extraction step involves extracting the temperature sampling values ​​and sampling time information corresponding to each location within a continuous time period from the original temperature sequence data.

[0170] The amplitude determination step involves comparing the highest and lowest temperature values ​​at the same location within a continuous time period to determine the amplitude of temperature change at that location within the continuous time period.

[0171] The rate determination step involves processing the temperature change between adjacent sampling times and the corresponding time interval to obtain the rate of temperature change in each location within a continuous time period.

[0172] The duration determination step involves statistically analyzing the length of time that the temperature in each direction continuously rises, continuously falls, or continuously maintains the same trend of change, and obtaining the corresponding duration of change.

[0173] The set generation step involves associating and organizing the temperature change amplitude, temperature change rate, and change duration corresponding to each location to generate a set of temperature change features corresponding to each location.

[0174] In a preferred embodiment of the present invention, the priority sequence generation subunit includes:

[0175] The feature reading step involves reading the temperature change feature set corresponding to each location and extracting the start time, magnitude, and duration of the change for each location within the same time period.

[0176] The steps for determining the sequence of temperature changes are compared to determine the order of temperature changes within the same time period.

[0177] The step of determining the intensity of change involves comprehensively comparing the temperature change amplitude, temperature change rate, and change duration in each direction to determine the strength of the influence of temperature change in each direction on the crystal oscillator temperature.

[0178] The priority sorting step sorts each location according to its order and the strength of its influence, giving higher priority to locations that occurred earlier and had a stronger influence.

[0179] The sequence generation step arranges the priority results corresponding to each position in chronological order to generate a competition relationship sequence with priority order.

[0180] In a preferred embodiment of the present invention, the competition state generation subunit includes:

[0181] The sequence reading step involves reading the priority results of each position corresponding to multiple consecutive time periods in the competition relationship sequence.

[0182] The state comparison step compares whether the highest priority location remains consistent within adjacent time periods and identifies whether the priority order has been adjusted.

[0183] The state determination step is as follows: when the highest priority position remains consistent and the priority order does not change, the competition relationship is determined to be a stable state; when the highest priority position changes or the priority order is adjusted, the competition relationship is determined to be a changing state.

[0184] The identifier generation step involves generating a corresponding competition status identifier based on the judgment result and establishing a correspondence between it and the time period to which it belongs.

[0185] The continuous impact recording step involves continuously recording the duration, intensity, and corresponding directional information of the dominant heating area's influence over subsequent time periods when the competitive relationship is stable, thus forming a continuous impact status record of the dominant heating area.

[0186] In a preferred embodiment of the present invention, the dominant switching determination unit includes:

[0187] The dominant region generation sub-unit is used to determine the dominant heating region in the current time period based on the competition status identifier and the highest priority orientation in the competition relationship sequence, and to generate the dominant region identifier.

[0188] The switching process generation subunit is used to identify the alternation process between the original dominant heating area and the new dominant heating area based on the change of the dominant area identifier between adjacent time periods, and to generate a dominant switching process sequence.

[0189] The switching identifier generation subunit is used to determine the corresponding time period when the original dominant heating area switches to the new dominant heating area based on the dominant switching process sequence, and to generate a dominant heat source switching identifier. At the same time, it records the residual impact information of the original dominant heating area after the switch, so as to be called in subsequent weight allocation.

[0190] In this embodiment of the invention, the dominant region generation subunit determines the dominant heating region in the current time period based on the competition status identifier and the highest priority position in the competition relationship sequence. This enables the heating sources that have a major impact on the crystal oscillator in different time periods to be clearly distinguished, thereby transforming the thermal impact analysis from an overall judgment to a process of identification for a specific position, providing a basis for subsequent dominant change analysis.

[0191] The switching process generation sub-unit continuously identifies the changes in the dominant region identifier in adjacent time periods, forming a process description of the transformation of the dominant heated region from its original location to a new location. This makes the change of the dominant heat source no longer represented by discrete results, but by a switching process with temporal continuity, thus reflecting the time of occurrence and the duration of the change in the dominant heat source.

[0192] The switching identifier generation subunit determines the specific time period during which the dominant heat source switches based on the dominant switching process sequence and generates the corresponding dominant heat source switching identifier. At the same time, it records the residual impact information of the original dominant heating area after the switch, so that the change of the dominant heat source can not only reflect the switching result, but also reflect the connection relationship of the impact before and after the switch. This provides a time location basis and residual impact basis for the subsequent weight allocation, enabling different heat sources to participate in the calculation of the crystal oscillator heating state in a continuous manner during the switching process.

[0193] In a preferred embodiment of the present invention, the dominant region generation subunit includes:

[0194] The status reading step involves reading the status result corresponding to the current time period from the competition status identifier, and simultaneously reading the priority ranking result of each position in the competition relationship sequence within that time period.

[0195] The candidate location extraction step extracts the location ranked first from the priority ranking results as the dominant candidate region.

[0196] The consistency verification step verifies whether the dominant candidate region remains consistent in consecutive adjacent time periods when the competition is in a stable state, and identifies the changing relationship between the dominant candidate region and the dominant region in the previous time period when the competition is in a changing state.

[0197] The area determination step involves identifying the location with the highest impact and meeting the status determination criteria within the current time period as the dominant heating area based on the consistency verification results.

[0198] The identifier generation step involves associating the identified dominant heating area with the corresponding time period to generate a dominant area identifier.

[0199] In a preferred embodiment of the present invention, the switching process generation subunit includes:

[0200] The identification reading steps involve reading the dominant region identifiers of multiple consecutive time periods in chronological order.

[0201] The adjacent comparison step compares the dominant region identifiers within adjacent time periods to identify whether the dominant heating region has changed from the original dominant heating region to a new dominant heating region.

[0202] The step of determining the start point of the handover is to determine the start time of the dominant handover as the moment when the dominant area identifier first changes in adjacent time periods.

[0203] Switch to continuous identification steps to track continuous time periods after the dominant region changes, and identify the temporal continuity between the weakening process of the original dominant heating region's influence and the strengthening process of the new dominant heating region's influence.

[0204] The switching endpoint determination step is to determine the end time of the dominant switching when the new dominant heating area continuously maintains the first priority and the original dominant heating area no longer occupies the dominant position.

[0205] The sequence generation step associates the start time of the dominant handover, the duration of the handover, and the end time of the handover in chronological order to generate a sequence of the dominant handover process.

[0206] In a preferred embodiment of the present invention, the switching identifier generation subunit includes:

[0207] The process sequence reading steps involve reading the original dominant heating area, the new dominant heating area, and the corresponding time period information in the dominant switching process sequence.

[0208] The steps for determining the switching period are as follows: based on the alternation of the weakening influence of the original dominant heating area and the strengthening influence of the new dominant heating area in the dominant switching process sequence, the corresponding time period when the dominant heat source switches is determined.

[0209] The identification generation step associates the original dominant heating area, the new dominant heating area, and the corresponding switching time period to generate a dominant heat source switching identification.

[0210] The residual information recording step involves recording the degree and duration of residual impact of the original dominant heating area after the switch, based on its continuous effect state before the switch, thus forming residual impact recording data.

[0211] The output step outputs the dominant heat source switching identifier and provides the residual impact record data to the dynamic weight allocation unit for invocation.

[0212] In a preferred embodiment of the present invention, the dynamic weight allocation unit includes:

[0213] The stage weight generation subunit is used to calculate the weights of the original dominant heating area, the new dominant heating area, and the residual influence area based on the dominant switching process sequence, the dominant heat source switching identifier, and the residual influence record data, and generate the corresponding stage effect weight sequence.

[0214] The continuous weight generation subunit is used to continuously adjust the corresponding weights of each direction according to the time-varying relationship of the phased weight sequence, so that the weight changes during the switching of the dominant heat source remain continuous and generate a weight distribution sequence that changes continuously over time.

[0215] The coupling weight generation subunit is used to couple the weights of each direction based on the mutual influence relationship between different directions in the weight distribution sequence and in combination with the dominant heat source switching identifier. This couples the weights of each direction to form a mutual constraint relationship between the original dominant heating area, the new dominant heating area, and the residual influence area, generating a composite weight sequence that reflects the combined effect of multiple directions. This composite weight sequence is then used as the weight corresponding to the dominant heat source identifier.

[0216] In this embodiment of the invention, the stage weight generation subunit performs weight calculation processing on the original dominant heating area, the new dominant heating area, and the residual influence area according to the dominant switching process sequence, the dominant heat source switching identifier, and the residual influence record data. This allows different heating areas to be distinguished according to their respective stages of action during the dominant switching process. As a result, the weight allocation is no longer determined by a single dominant result, but can simultaneously reflect the comprehensive relationship of the original dominant attenuation, the new dominant enhancement, and the continuation of the residual influence.

[0217] The continuous weight generation subunit continuously adjusts the weights of each direction according to the temporal relationship of the phase action weight sequence, so that different heated areas gradually transition from the original weight to the new weight state during the dominant switching process, thereby avoiding abrupt changes in weight between adjacent time periods and keeping the description of the crystal oscillator's heating state continuously changing in the time dimension.

[0218] The coupling weight generation subunit couples the weights of each direction based on the mutual influence relationship between different directions in the weight distribution sequence and in combination with the dominant heat source switching identifier. This makes the weight changes between the original dominant heating region, the new dominant heating region, and the residual influence region interrelated, so that the weights of each direction no longer change independently, but can reflect the comprehensive influence result under the joint action of multiple directions. In this way, the expression of the crystal oscillator heating state can reflect the synergistic effect process between different heat sources.

[0219] In a preferred embodiment of the present invention, the stage weight generation subunit includes:

[0220] The sequence reading step involves reading the information of the original dominant heating area and the new dominant heating area corresponding to each time period in the dominant switching process sequence, and reading the residual influence degree of the corresponding time period in the residual heat influence sequence.

[0221] The phase identification step involves identifying the dominant attenuation phase, dominant enhancement phase, and dominant alternation phase based on the dominant switching process sequence, and establishing a correlation between each phase and its corresponding time period.

[0222] The basic impact determination steps are to determine the impact degree of the original dominant heating area, the new dominant heating area, and the residual impact area in each stage. Specifically, the impact degree of the original dominant heating area decreases step by step in the dominant attenuation stage, the impact degree of the new dominant heating area increases step by step in the dominant enhancement stage, and the impact degree of the residual impact area is determined according to the residual thermal impact sequence in the corresponding time period.

[0223] The weighting process involves assigning corresponding weights to the three regions based on the proportion of the influence of the original dominant heating region, the new dominant heating region, and the residual influence region in the overall influence within the same time period.

[0224] The sequence generation step involves arranging the weight results corresponding to the original dominant heating area, the new dominant heating area, and the residual influence area in each time period in chronological order to generate a phase effect weight sequence.

[0225] In a preferred embodiment of the present invention, the continuous weight generation subunit includes:

[0226] The weight reading step reads the regional weight results corresponding to each time period in the weight sequence of the stage effect; the adjacent comparison step compares the weight changes of the same region in adjacent time periods to identify the direction and magnitude of weight changes.

[0227] The transition interval setting steps involve setting a weight transition interval between adjacent time periods, so that the weight result of the previous time period gradually transitions to the weight result of the next time period.

[0228] The continuous adjustment process involves adjusting the weights of the original dominant heating region, the new dominant heating region, and the residual influence region in stages according to the time progression within the weight transition interval, so that the weight change process of each region remains continuous.

[0229] The distribution association step establishes a correlation between the continuously adjusted weights of each region and the corresponding time period and location.

[0230] The sequence generation step involves arranging the continuously adjusted regional weights within each time period in chronological order to generate a weight distribution sequence that changes continuously over time.

[0231] In a preferred embodiment of the present invention, the coupling weight generation subunit includes:

[0232] The steps for reading weighted data are as follows: read the corresponding influence weights of each location within the same time period;

[0233] The correlation identification step identifies the mutual influence relationships between different locations based on the adjacency of each location in space and the consistency of the direction of temperature change.

[0234] The steps for determining the dominant relationship are as follows: among multiple directions that have mutual influence relationships, determine the direction that plays a dominant role and its related directions;

[0235] The weight adjustment process involves compensating for changes in the weights of related locations by adjusting the weights of the dominant location and simultaneously adjusting the weights of related locations based on changes in the weights of the dominant location, thus creating a linkage between the weights of each location.

[0236] The composite generation step summarizes the adjusted weights of each aspect to generate a composite weight sequence that reflects the combined effect of multiple aspects.

[0237] In a preferred embodiment of the present invention, the path difference parsing unit includes:

[0238] The path time series generation subunit is used to extract the start and end times of temperature changes in each path based on the temperature change process in each path over a continuous time period, and generate the corresponding path time series data.

[0239] The delay interval generation sub-unit is used to perform interval calculation processing on the temperature transfer process of each path based on the time difference between the start and end times of the temperature change of each path in the path time series data, and generate the corresponding path delay interval.

[0240] The state association correction subunit is used to perform state association correction processing on the path delay interval according to the temperature change amplitude and trend during the temperature change process of each path, so that the delay interval of each path is adjusted with the temperature change state, and the independent transmission delay interval corresponding to each path is determined.

[0241] In this embodiment of the invention, the path timing generation subunit extracts the start and end times of temperature changes for each path to form path timing data, enabling the temperature change process of each path to be expressed in the time dimension; the delay interval generation subunit calculates the temperature transmission time difference for each path based on the path timing data to form path delay intervals, thus distinguishing the transmission characteristics of different paths; the state association correction subunit adjusts the path delay intervals according to the temperature change amplitude and trend, so that the delay intervals can change with the temperature change state, and finally determines the independent transmission delay intervals corresponding to each path, thereby enabling the delay characteristics to reflect the changes in the actual transmission process.

[0242] In a preferred embodiment of the present invention, the path time series generation subunit is used to extract the start and end times of temperature changes in each path based on the temperature change process within a continuous time period, and generate corresponding path time series data, including:

[0243] The path data reading steps involve reading the temperature sampling values, sampling times, and path identification information corresponding to each path within a continuous time period.

[0244] The step of identifying the starting point of change involves analyzing the temperature sampling values ​​arranged in chronological order within the same path. The moment when the temperature changes from a stable state to a continuous rise or a continuous fall with a continuously increasing amplitude is determined as the starting moment of the temperature change along that path.

[0245] The change endpoint identification step continuously tracks the temperature change process. When the temperature change amplitude decreases and the rate of temperature change remains within a preset range for multiple consecutive sampling periods, the corresponding sampling time is determined as the end time of the temperature change.

[0246] The time period correspondence step associates the start and end times of each path with the corresponding path identifier, so that each path forms an independent time segment;

[0247] The data generation step involves summarizing the time segments corresponding to each path according to the path, and generating the corresponding path time series data.

[0248] In a preferred embodiment of the present invention, a delay interval generation subunit is used to determine the time delay of the thermal impact of each path from the dominant heated region to the crystal oscillator position based on path timing data, and to generate the corresponding path delay interval, including:

[0249] The reference time determination step involves reading the moment when the temperature change of the dominant heated region begins as the reference start time for the generation of thermal effects.

[0250] The path response identification step analyzes the temperature change of each path at the crystal oscillator position and determines the response time of the path when the temperature at the crystal oscillator position begins to change accordingly.

[0251] The delay time determination step determines the time interval between the response time of each path and the reference start time, thus obtaining the time delay experienced by the thermal effects of that path to the crystal oscillator position.

[0252] The interval construction step uses the time range between the baseline start time and the path response time as the delay interval of the path;

[0253] The interval generation step associates the delay intervals corresponding to each path with the path identifier to generate the corresponding path delay intervals.

[0254] In a preferred embodiment of the present invention, the state association correction subunit is used to perform state association correction processing on the path delay interval based on the temperature change amplitude and rate of change during the temperature change process of each path, and to determine the independent transmission delay interval corresponding to each path, including:

[0255] The status parameter reading steps include reading the temperature change amplitude, temperature change direction, and temperature change rate of each path within a continuous time period, and reading the corresponding path delay interval.

[0256] The state change identification step compares the temperature change status of the same path in different time periods to identify situations where the temperature change accelerates, slows down, or changes in the direction of change.

[0257] The interval position adjustment steps are as follows: when the rate of temperature change increases, the path response time is adjusted forward, causing the entire delay interval to move forward; when the rate of temperature change decreases, the path response time is adjusted backward, causing the entire delay interval to move backward.

[0258] The interval range correction step adjusts the length of the delay interval accordingly based on the increase or decrease of the temperature change amplitude, so that the delay interval can reflect the continuous process of heat effect transmission.

[0259] The independent interval determination step involves marking the corrected path delay intervals separately to determine the independent transmission delay intervals corresponding to each path.

[0260] In a preferred embodiment of the present invention, the delay competition superposition unit includes:

[0261] The arrival order generation sub-unit is used to sort the time order of the thermal effects of different paths to the crystal oscillator position according to the independent transmission delay interval, and generate the path arrival order sequence.

[0262] The state transfer generation sub-unit is used to generate a state transfer sequence between paths by taking the temperature state formed at the crystal oscillator position of the first arriving path as the initial state of the thermal influence of the later arriving path, according to the path arrival sequence.

[0263] The nonlinear superposition generation sub-unit is used to successively superimpose the thermal effects of different paths on the basis of the previous state according to the state transfer sequence between paths, so that the effect of the later arriving path depends on the temperature state formed by the earlier arriving path, and generates a local thermal effect response sequence containing the interdependence between paths.

[0264] In this embodiment of the invention, the arrival sequence generation subunit sorts the arrival time of the thermal effects of different paths at the crystal oscillator position according to the independent transmission delay interval, so that the action order of each path is clear; the state transfer generation subunit takes the temperature state formed by the first arriving path at the crystal oscillator position as the initial action state of the subsequent arriving path, so that a state transfer relationship is formed between different paths; the nonlinear superposition generation subunit superimposes the thermal effects of each path on the basis of the state transfer relationship, so that the action result of the subsequent path depends on the previous state, thereby enabling the local thermal effect response sequence to reflect the interdependent influence process between paths.

[0265] In a preferred embodiment of the present invention, the arrival sequence generation subunit includes:

[0266] The interval data reading step involves reading the independent transmission delay interval corresponding to each path and extracting the start time, end time, and path identification information of each path delay interval.

[0267] The sequence comparison step compares the start times of each path delay interval to determine the order in which the thermal effects of different paths begin to reach the crystal oscillator position. If the start times are the same, the end times of the corresponding delay intervals are compared to determine the order in which the continuous effect ends first or last.

[0268] The sequential numbering step assigns sequential numbers to each path according to the comparison results, so that the path that arrives at the crystal oscillator position first has a higher priority, and the subsequent paths are arranged in order.

[0269] The same-order path differentiation step involves further differentiating these multiple paths based on the duration of their corresponding thermal effects and the intensity of temperature changes at the initial stage of their action when multiple paths arrive at the crystal oscillator position at the same time, thereby establishing an internal sorting relationship between the same-order arrival paths.

[0270] The sequence generation step involves summarizing the path identifiers, corresponding sequence numbers, and arrival order of each path in chronological order to generate a path arrival order sequence.

[0271] In a preferred embodiment of the present invention, the state transfer generation subunit includes:

[0272] The sequential sequence reading step involves reading the arrival order information of each path in the path arrival sequence and determining the relationship between the current path and the previous path.

[0273] The preceding state extraction step extracts the temperature state formed at the crystal oscillator position by the first path to arrive. The temperature state includes the temperature level at the end of the path's operation, the direction of temperature change, and the duration of temperature change.

[0274] The initial state setting step is to determine the temperature state formed by the first arriving path as the initial operating state of the subsequent arriving path when the subsequent arriving path begins to operate, so that the subsequent arriving path does not use the standard initial temperature as the starting point of operation, but uses the temperature state already formed by the preceding path as the basis for operation.

[0275] In the state connection step, for the case where multiple paths arrive in sequence, the temperature state formed by the preceding path is transmitted segment by segment according to the path arrival order, so that the starting action state of each path is corresponding to the ending action state of the preceding path.

[0276] The sequence generation step involves sequentially associating and organizing the preceding temperature states, the initial action states of the current path, and their corresponding time periods for each path to generate a state transfer sequence between paths.

[0277] In a preferred embodiment of the present invention, a nonlinear superposition generation subunit is used to perform nonlinear superposition processing on the thermal effects of different paths based on the previous state according to the state transfer sequence between paths, generating a local thermal effect response sequence containing the interdependence between paths, including:

[0278] The sequence reading step involves reading the initial action state and thermal impact content of each path in the inter-path state transfer sequence.

[0279] The basic state determination step is to take the initial action state corresponding to the current path as the basic temperature state for the thermal effect superposition of the path.

[0280] The adjustment steps adjust the degree of thermal influence of the current path according to the base temperature. When the base temperature is high, the temperature increment of the current path is reduced, and when the base temperature is low, the temperature increment of the current path is increased.

[0281] The nonlinear superposition step superimposes the adjusted thermal effect with the base temperature state to form a new temperature state after the current path effect.

[0282] The state transfer step takes the temperature state after the current path is applied as the base state for the next path to be superimposed, so that the effect of the subsequent path depends on the previous state.

[0283] The response sequence generation step arranges the temperature states and their dependencies formed by each path within the corresponding time period in chronological order to generate a local thermal effect response sequence containing nonlinear superposition relationships.

[0284] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A system for calibrating the accuracy of a timepiece driven by an intelligent sensor, characterized in that, The system includes: The spatial orientation acquisition module is used to deploy temperature sensing units in different heated orientations of the clock timing device, collect the spatial position of each temperature sensing unit and the corresponding real-time temperature data, and generate raw temperature sequence data with time stamps and orientation identifiers. The heat source dominance identification module is used to calculate the rate and duration of temperature change in each direction based on the original temperature sequence data, identify the dominant heated area that has the main impact on the crystal oscillator, and generate the corresponding dominant heat source identifier and its influence weight. The heat transfer delay modeling module is used to analyze the temperature change transfer time difference between the dominant heat source and the crystal oscillator position based on the dominant heat source identification and crystal oscillator installation position, and construct a local thermal effect response sequence that includes time lag relationships. The predicted frequency offset generation module is used to deduce the frequency evolution sequence of the crystal oscillator in the subsequent time period based on the local thermal effect response sequence, and generate the corresponding predicted frequency offset. The feedforward calibration control module is used to pre-adjust the cumulative counting rhythm of the timing pulses before the timing error actually occurs, based on the predicted frequency offset, and generate a feedforward corrected timing reference signal. The closed-loop adaptive correction module is used to compare the timing reference signal with an external standard time source and correct the dominant heat source weight and heat transfer delay relationship to continuously optimize prediction accuracy.

2. The intelligent sensor-driven clock timing accuracy calibration system according to claim 1, characterized in that, The heat source dominance identification module includes: The multi-source competitive evaluation unit is used to jointly analyze the magnitude, rate, and duration of temperature changes in each location within a continuous time period in the original temperature sequence data, forming a competitive relationship sequence of the influence of temperature changes in each location on the crystal oscillator temperature. The dominant switching determination unit is used to identify the dominant heated area in the current time period in the competition relationship sequence, and generate a dominant heat source switching identifier when the degree of influence between different directions alternates, and record the corresponding switching time period. The dynamic weight allocation unit is used to assign a time-varying weight to the corresponding dominant heat source identifier based on the continuity and superposition of the influence of each direction before and after the switch during the process of switching the dominant heat source.

3. The intelligent sensor-driven clock timing accuracy calibration system according to claim 1, characterized in that, The heat transfer delay modeling module includes: The multi-path partitioning unit is used to divide the heat transfer process from the dominant heated area to the crystal oscillator position into multiple paths with different structures based on the spatial structure between the dominant heat source identifier and the crystal oscillator installation position. The path difference analysis unit is used to analyze the start time and completion time of temperature change transmission in each path, and determine the independent transmission delay interval for each path. The delay competition superposition unit is used to superimpose the thermal effects of different paths in time according to the propagation delay interval of each path and the order in which they arrive at the crystal oscillator position, forming a local thermal effect response sequence that includes the sequential action relationship of multiple paths.

4. The intelligent sensor-driven clock timing accuracy calibration system according to claim 1, characterized in that, The predicted frequency offset generation module includes: The delay chain expansion unit is used to expand the crystal oscillator temperature change process in stages according to the arrival order of thermal effects along different paths in the local thermal effect response sequence, forming a temperature effect evolution sequence containing multiple time periods. The stage superposition derivation unit is used to perform stage-by-stage superposition derivation of the crystal oscillator frequency change process according to the sequential relationship of each stage in the temperature influence evolution sequence, forming a frequency evolution sequence that reflects the frequency change with time in the future time period. The forward offset determination unit is used to determine the predicted frequency offset for the corresponding time period based on the frequency change portion of the corresponding future time period in the frequency evolution sequence.

5. The intelligent sensor-driven clock timing accuracy calibration system according to claim 1, characterized in that, The feedforward calibration control module includes: The offset distribution generation unit is used to segment and expand the frequency changes in future time periods according to the predicted frequency offset and in chronological order, forming a frequency offset distribution sequence for each time period. The rhythm mapping generation unit is used to map the frequency offset of each time period to the adjustment rhythm of the timing pulse cumulative count according to the frequency offset distribution sequence, and generate the corresponding counting rhythm control sequence. The pre-adjustment execution unit is used to pre-adjust the cumulative counting process of timing pulses before the corresponding time period arrives according to the counting rhythm control sequence, and generate an initial timing reference signal corresponding to the frequency offset distribution sequence. The continuous constraint correction unit is used to perform continuous constraint processing on the initial timing reference signal according to the rhythm change relationship between adjacent time periods in the counting rhythm control sequence, correct the transition of counting changes between adjacent time periods, and generate a continuously changing timing reference signal.

6. The intelligent sensor-driven clock timing accuracy calibration system according to claim 2, characterized in that, The multi-source competition evaluation unit includes: The feature set generation subunit is used to correlate and combine the magnitude, rate and duration of temperature change in each location in the original temperature sequence data over a continuous time period to generate a temperature change feature set corresponding to each location. The priority sequence generation subunit is used to sort the influence of temperature changes on crystal oscillator temperature in each direction according to the order and intensity of changes in each direction within the same time period in the temperature change feature set, and generate a competitive relationship sequence with priority order. The competition state generation subunit is used to determine whether the competition relationship is in a stable or changing state based on the change of priority of each position in the competition relationship sequence over time, generate the corresponding competition state identifier, and record the continuous influence state of the dominant heated area in the stable state in the subsequent time period.

7. The intelligent sensor-driven clock timing accuracy calibration system according to claim 6, characterized in that, The dominant handover determination unit includes: The dominant region generation sub-unit is used to determine the dominant heating region in the current time period based on the competition status identifier and the highest priority orientation in the competition relationship sequence, and to generate the dominant region identifier. The switching process generation subunit is used to identify the alternation process between the original dominant heating area and the new dominant heating area based on the change of the dominant area identifier between adjacent time periods, and to generate a dominant switching process sequence. The switching identifier generation subunit is used to determine the corresponding time period when the original dominant heating area switches to the new dominant heating area based on the dominant switching process sequence, and to generate a dominant heat source switching identifier. At the same time, it records the residual impact information of the original dominant heating area after the switch, so as to be called in subsequent weight allocation.

8. The intelligent sensor-driven clock timing accuracy calibration system according to claim 7, characterized in that, The dynamic weight allocation unit includes: The stage weight generation subunit is used to calculate the weights of the original dominant heating area, the new dominant heating area, and the residual influence area based on the dominant switching process sequence, the dominant heat source switching identifier, and the residual influence record data, and generate the corresponding stage effect weight sequence. The continuous weight generation subunit is used to continuously adjust the corresponding weights of each direction according to the time-varying relationship of the phased weight sequence, so that the weight changes during the switching of the dominant heat source remain continuous and generate a weight distribution sequence that changes continuously over time. The coupling weight generation subunit is used to couple the weights of each direction based on the mutual influence relationship between different directions in the weight distribution sequence and in combination with the dominant heat source switching identifier. This couples the weights of each direction to form a mutual constraint relationship between the original dominant heating area, the new dominant heating area, and the residual influence area, generating a composite weight sequence that reflects the combined effect of multiple directions. This composite weight sequence is then used as the weight corresponding to the dominant heat source identifier.

9. The intelligent sensor-driven clock timing accuracy calibration system according to claim 3, characterized in that, The path difference parsing unit includes: The path time series generation subunit is used to extract the start and end times of temperature changes in each path based on the temperature change process in each path over a continuous time period, and generate the corresponding path time series data. The delay interval generation sub-unit is used to perform interval calculation processing on the temperature transfer process of each path based on the time difference between the start and end times of the temperature change of each path in the path time series data, and generate the corresponding path delay interval. The state association correction subunit is used to perform state association correction processing on the path delay interval according to the temperature change amplitude and trend during the temperature change process of each path, so that the delay interval of each path is adjusted with the temperature change state, and the independent transmission delay interval corresponding to each path is determined.

10. The intelligent sensor-driven clock timing accuracy calibration system according to claim 9, characterized in that, The delay competition superposition unit includes: The arrival order generation sub-unit is used to sort the time order of the thermal effects of different paths to the crystal oscillator position according to the independent transmission delay interval, and generate the path arrival order sequence. The state transfer generation sub-unit is used to generate a state transfer sequence between paths by taking the temperature state formed at the crystal oscillator position of the first arriving path as the initial state of the thermal influence of the later arriving path, according to the path arrival sequence. The nonlinear superposition generation sub-unit is used to successively superimpose the thermal effects of different paths on the basis of the previous state according to the state transfer sequence between paths, so that the effect of the later arriving path depends on the temperature state formed by the earlier arriving path, and generates a local thermal effect response sequence containing the interdependence between paths.

Images