Intelligent electric control system and control circuit board with same

Abstract

The invention relates to the technical field of intelligent electric control systems, and discloses an intelligent electric control system and a control circuit board with the intelligent electric control system. The system comprises a signal acquisition module used for acquiring an input signal and judging a signal partition; the mode mapping module is used for constructing a mapping relation between the signal partition and the control partition; the parameter combination module is used for extracting key partitions and judging a parameter combination matching degree to obtain a bias combination; the state updating module is used for updating the mapping relation based on the deviation combination; and the control management module generates a control management library. The control circuit board integrates each module hardware circuit unit to realize signal transmission and data interaction. Accurate processing, dynamic mapping, optimal parameter combination, state real-time updating and scientific control management of the input signals are achieved, the intelligence, adaptability and reliability of the electronic control system are improved, and the system is suitable for electronic control scenes of various intelligent devices.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent electric control systems, and in particular to an intelligent electric control system and a control circuit board having the intelligent electric control system. Background Art

[0002] In the development of modern industrial automation and intelligence, electronic control systems, as the core control component of equipment operation, have a direct impact on the performance and efficiency of their overall operational effectiveness. As equipment functions become increasingly complex and diverse, higher requirements are placed on the intelligent and precise control of electronic control systems. Traditional electronic control systems often suffer from incomplete signal processing and simplistic control logic when faced with complex input signals. For example, after acquiring input signals during equipment operation, it is difficult to accurately determine the signal partitioning of the input signal relative to the electronic control function, resulting in an inability to fully understand the signal characteristics and properties, which in turn affects subsequent control decisions.

[0003] In terms of signal classification, traditional systems lack effective type recognition models and classification methods, cannot accurately classify input signals, and cannot obtain comprehensive type categories. It is also difficult to identify high-frequency features, ambiguous features, and correlation features in the signals, and cannot form a complete signal feature set, resulting in insufficient depth and breadth of signal processing.

[0004] When it comes to analyzing and partitioning signal features, traditional methods do not fully consider factors such as feature merging, comparison, change rate, and intensity. This makes it difficult to accurately obtain high-frequency feature areas, ambiguous feature resolution areas, and associated feature association areas. This results in inaccurate signal partitioning and an inability to accurately reflect the actual needs of the equipment.

[0005] In terms of pattern mapping, traditional systems cannot effectively call the parameters and states corresponding to the input signal, generate unlabeled parameter identification results, and determine whether they are the target parameter identification results, making it difficult to determine the control partition of the input signal, affecting the construction of the mapping relationship between signal partition and control partition.

[0006] In traditional electronic control systems, the parameter combination module lacks effective clustering analysis of signal partitions and control partitions, making it difficult to determine key partitions, extract key features and calculate feature similarity, set common sequences and longest common subsequences, resulting in the inability to accurately obtain the combination matching degree and parameter bias combination between parameters, affecting the accuracy of control.

[0007] In traditional systems, the state update module cannot set the target path based on the time distribution probability of each parameter in the parameter bias combination, fit the target path and obtain the probability of combination occurrence, compare the difference with the reference status of the state in the control partition, and complete the state update, resulting in poor system adaptability and real-time performance.

[0008] In traditional electronic control systems, the control management module cannot accurately extract control objectives based on updated mapping relationships, derive control sequences by probability of occurrence, and combine them into a control management library. This leads to unclear control objectives and a chaotic control sequence, impacting the normal operation of the equipment. When integrating the various modules of the electronic control system on traditional control circuit boards, signal transmission and data exchange between hardware circuit units are not efficient and stable, affecting the performance of the entire electronic control system. Summary of the Invention

[0009] The object of the present invention is to provide an intelligent electronic control system and a control circuit board having the intelligent electronic control system to solve the problems raised in the above background technology.

[0010] To achieve the above objectives, the present invention provides the following technical solutions: an intelligent electronic control system and a device having the intelligent electronic control system, the system comprising: The signal acquisition module is used to obtain the input signal when the equipment is running, and determine the signal partition of the input signal relative to the electronic control function according to the control requirements of the equipment during operation; A pattern mapping module is used to obtain the control partition of the input signal and build a mapping relationship between the signal partition and the control partition; The parameter combination module is used to extract key partitions from the signal partition and the control partition, and judge the combination matching degree between multiple parameters in the control partition according to the key partition to obtain the parameter bias combination; A state update module is used to verify the information of the control partition based on the parameter bias combination, identify the update of the state in the control partition, and update the mapping relationship between the signal partition and the control partition according to the update of the state; The control management module is used to identify the control target and control sequence of the input signal according to the updated mapping relationship between the signal partition and the control partition, and generate a control management library.

[0011] Preferably, the signal acquisition module is implemented as follows: For any input signal during device operation, obtain a type recognition model corresponding to the input signal; Classify the input signal using a type recognition model to obtain at least one type category; Identify high-frequency features, ambiguous features, and related features in the input signal under the corresponding type category to form a signal feature set; The high-frequency features, ambiguous features and associated features in the signal feature set are analyzed respectively, and the high-frequency feature area, ambiguous feature resolution area and associated feature association area corresponding to the signal feature set are obtained in turn, and used as the signal partitioning of the input signal relative to the electronic control function.

[0012] Preferably, the implementation method of obtaining the high-frequency feature area, the ambiguous feature resolution area, and the associated feature association area corresponding to the signal feature set further includes: Merge the high-frequency features, ambiguous features, and associated features in the signal feature set according to type categories to obtain multiple type merging results; Extract high-frequency feature pairs from the type merging results, compare the high-frequency feature pairs with the feature library, and obtain high-frequency feature areas; The resolution change rate of the ambiguous features and the association strength of the associated features in the type merging result are extracted, and the type merging result is divided according to the resolution change rate of the ambiguous features and the association strength of the associated features to obtain the ambiguous feature resolution area and the associated feature association area.

[0013] Preferably, the implementation method of determining the signal partition of the input signal relative to the electronic control function further includes: Judge the signal partitions, analyze the signal times and signal times of the devices in the signal partitions, fit the signal partitions according to the signal times and signal times, and build a mapping relationship between the signal partitions and the actual needs of the devices.

[0014] Preferably, the implementation of the mode mapping module includes: Call the parameters and states corresponding to the input signal to generate multiple unlabeled parameter recognition results. The unlabeled parameter recognition results represent the parameters and states that are not associated with the features in the signal partition; It is determined whether the multiple unlabeled parameter recognition results are target parameter recognition results. If they are target parameter recognition results, the target parameter recognition results are regarded as the control partitions of the input signal.

[0015] Preferably, the implementation method of constructing the mapping relationship between the signal partition and the control partition includes: using the information representation of the high-frequency feature area, ambiguous feature resolution area and associated feature association area existing in the signal partition, the descriptive information of the control partition and the category of the control partition to construct a mapping relationship between the signal partition and the control partition.

[0016] Preferably, the parameter combination module is implemented as follows: Perform cluster analysis on signal partitions and control partitions according to control type, function type, and execution function, and set the largest cluster center after cluster analysis as the key partition; Extracting key features of key partitions, calculating feature similarities between key features, and setting a common sequence related to the feature similarities between key features; Using the common sequence related to the feature similarity between the key features, the parameters existing in the common sequence are extracted, and the longest common subsequence between the parameters is set, and the length value of the longest common subsequence is set as the combined matching degree between the parameters; The combination matching degree between the parameters is set according to the time distribution probability of each parameter, and the parameter bias combination is set.

[0017] Preferably, the implementation of the status update module includes: Extract the time distribution probability of each parameter from the parameter bias combination; set the target path of the parameter bias combination according to the time period corresponding to the time distribution probability of each parameter; Fitting the target path of each parameter in the parameter bias combination to obtain a fitted target path, and setting the probability value of the fitted target path in each time period as the combined occurrence probability of the parameter bias combination; The probability of occurrence of the parameter bias combination is compared with the reference of the state in the control partition, the difference value is identified, and the state in the control partition is classified according to the difference value, thereby completing the update of the state in the control partition.

[0018] Preferably, the control management module is implemented as follows: According to the mapping relationship between the updated signal partition and the control partition, the control targets in the input signal are extracted, and the control targets are sorted according to the probability of occurrence of the control targets to obtain the control sequence in the input signal; the control targets and the control sequence are combined in a structured form to obtain a control management library.

[0019] Preferably, the present invention further comprises a control circuit board having the above-mentioned intelligent electronic control system, wherein the circuit board further comprises: A hardware circuit unit integrating the signal acquisition module, mode mapping module, parameter combination module, state update module and control management module realizes signal transmission and data interaction between modules through circuit connection.

[0020] Compared with the prior art, the present invention has the following beneficial effects: The intelligent electronic control system and control circuit board provided by the present invention have significant beneficial effects in many aspects. The signal acquisition module obtains the type recognition model corresponding to the input signal, classifies the input signal and identifies high-frequency, ambiguous and related features to form a signal feature set, and then analyzes and obtains signal partitions. It can comprehensively and accurately process the input signal, deeply understand the signal characteristics, and provide a reliable basis for subsequent control. The pattern mapping module calls parameters and states to generate unlabeled parameter recognition results, determines whether it is the target parameter recognition result to determine the control partition, and constructs a mapping relationship between signal partitions and control partitions, making the correspondence between signal and control clearer and improving the accuracy of control.

[0021] The parameter combination module performs cluster analysis on signal and control partitions to identify key partitions, extracts key features, calculates similarity, and sets common sequences and longest common subsequences to obtain biased parameter combinations. This module accurately analyzes the matching degree between parameter combinations, optimizes parameter combinations, and improves control accuracy and rationality. The state update module sets the target path based on the temporal distribution probability of the biased parameter combinations. After fitting, the probability of the combination is obtained and compared with the state reference to complete the state update. This enables the system to adjust its state in real time according to actual conditions, enhancing its adaptability and real-time performance.

[0022] The control management module extracts control targets based on the updated mapping relationships and sorts them into a control sequence, which is then combined into a control management library. This clarifies the control targets and sequence, making equipment operation more orderly and efficient. The control circuit board with this intelligent electronic control system integrates the hardware circuit units of each module to achieve efficient signal transmission and data exchange between modules, ensuring stable and reliable operation of the electronic control system and improving overall performance.

[0023] This intelligent electronic control system and control circuit board work closely together and collaborate with each other in all aspects, from signal acquisition and processing to control management, which comprehensively improves the intelligence level, control accuracy, adaptability and real-time performance of the electronic control system, meets the complex and diverse control needs of modern equipment, and has broad application prospects and significant practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 The intelligent electronic control system of the present invention and the working principle diagram of the intelligent electronic control system; Figure 2 This is the design diagram of the signal acquisition module; Figure 3 Design diagram for the pattern mapping module; Figure 4 This is the design diagram of the status update module. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0026] See also Figures 1-4 The present invention relates to an intelligent electronic control system and a control circuit board having the intelligent electronic control system. The system includes: a signal acquisition module, a mode mapping module, a parameter combination module, a status update module, and a control management module. The specific implementation steps are as follows: The signal acquisition module is used to obtain the input signal when the equipment is running, and determine the signal partition of the input signal relative to the electronic control function according to the control requirements when the equipment is running.

[0027] The pattern mapping module is used to obtain the control partition of the input signal and build a mapping relationship between the signal partition and the control partition.

[0028] The parameter combination module is used to extract key partitions from the signal partition and the control partition, and judge the combination matching degree between multiple parameters in the control partition according to the key partition to obtain the parameter bias combination.

[0029] The state update module is used to verify the information of the control partition based on the parameter bias combination, identify the update of the state in the control partition, and update the mapping relationship between the signal partition and the control partition according to the update of the state.

[0030] The control management module is used to identify the control target and control sequence of the input signal according to the updated mapping relationship between the signal partition and the control partition, and generate a control management library.

[0031] Example 1: The implementation process of the signal acquisition module needs to be closely centered around the processing of input signals when the device is running. For any input signal when the device is running, the system obtains a type recognition model corresponding to the input signal. This model can be constructed based on deep learning or traditional machine learning algorithms, such as through training of convolutional neural networks or support vector machines. Its core function is to accurately classify the types of input signals. When the type recognition model is used to process the input signal, the model will output at least one type category based on the preset feature extraction rules and classification logic. For example, the input signal will be classified into different types such as temperature signals, pressure signals, and speed signals, laying the foundation for subsequent feature analysis.

[0032] After determining the type category of the input signal, the system needs to identify the high-frequency features, ambiguous features, and associated features present in the input signal under this type category to form a signal feature set. Among them, high-frequency features refer to features that appear frequently and are representative in this type of signal, such as the fluctuation frequency in a specific temperature range or the rate of change of a specific temperature in a temperature signal; ambiguous features refer to features that may lead to multiple possibilities for signal interpretation, such as certain interference signals or signal features under critical conditions; associated features are features that have a mutual influence or correlation relationship with other features, such as the correlation characteristics between temperature signals and pressure signals under specific working conditions. The recognition process of these features needs to be combined with the time domain and frequency domain analysis methods of the signal, and extracted through signal processing methods such as filtering and spectrum transformation.

[0033] After forming a signal feature set, the high-frequency, ambiguous, and associated features are analyzed separately. The corresponding high-frequency feature regions, ambiguous feature resolution regions, and associated feature association regions are sequentially obtained to serve as signal partitions of the input signal relative to the electronic control function. To obtain these partitions, the high-frequency, ambiguous, and associated features in the signal feature set are first merged according to type category, resulting in multiple type-merged results. For example, for the temperature signal category, all high-frequency, ambiguous, and associated features are merged into a single set to facilitate subsequent processing.

[0034] High-frequency feature pairs are extracted from the type merging results. This involves pairing together correlated or synergistic high-frequency features to form feature pairs. These high-frequency feature pairs are then compared with a pre-established feature library, which stores information about various known high-frequency feature pairs and their corresponding feature regions. This comparison identifies the high-frequency feature regions. For example, if the feature library records a feature pair for a temperature change rate and temperature fluctuation range that falls within a high-temperature warning high-frequency feature region, then if the input signal's feature pair matches this record, it can be assigned to the corresponding high-frequency feature region.

[0035] To process ambiguous and associated features, it is necessary to extract the resolution change rate of ambiguous features and the association strength of associated features in the type merging results. The resolution change rate is used to measure the degree of change in the resolution results of ambiguous features under different working conditions or time points. It can be obtained by calculating the variance or standard deviation of the feature resolution results. The association strength is used to characterize the degree of mutual influence between associated features and can be quantified using indicators such as the correlation coefficient. The type merging results are divided according to the resolution change rate of ambiguous features and the association strength of associated features. Ambiguous features with higher resolution change rates are divided into ambiguous feature resolution zones, and associated features with higher association strengths are divided into associated feature association zones.

[0036] When determining the signal partition of the input signal relative to the electronic control function, an in-depth analysis of the signal partition is also required. Specifically, the number of signals and signal time of the equipment in the signal partition are analyzed. The number of signals refers to the frequency with which the equipment receives the signal corresponding to the signal partition within a specific time period, and the signal time refers to the duration of the signal or the time point of occurrence. The signal partition is fitted according to the number of signals and signal time. Linear fitting, polynomial fitting or other suitable fitting methods can be used to construct a mapping relationship from the signal partition to the actual needs of the equipment. For example, when the number of signals in a certain signal partition is frequent and the signal time is long, the actual need for emergency regulation of the corresponding equipment can be determined through fitting analysis, thereby providing a basis for subsequent control strategies.

[0037] Example 2: The mode mapping module's workflow centers on processing input signal parameters and states, aiming to establish a precise mapping relationship between signal partitions and control partitions. This module calls the parameters and states corresponding to the input signal. Parameters include physical properties such as voltage amplitude, current intensity, and temperature, while states include the device's operating mode (e.g., standby, operating, fault) and signal transmission status (e.g., normal, interrupted, attenuated). This call process is implemented via the system bus or data interface, ensuring the real-time and integrity of parameter and state information.

[0038] After calling the parameters and states, the system generates multiple unlabeled parameter recognition results. These results are preliminary analyses of the input signal parameters and states, and have not yet established a correspondence with the high-frequency feature area, ambiguous feature resolution area, or associated feature association area in the signal partition. For example, when the input signal is the current parameter (10A) and operating state (high-speed operation) of a motor, the generated unlabeled parameter recognition results may include independent parameter items such as "current value 10A" and "operating speed level 3." At this time, these parameter items are not associated with any features in the signal partition.

[0039] It is necessary to determine whether multiple unlabeled parameter identification results are target parameter identification results. The judgment criteria are based on the system's preset control logic and equipment operation requirements, and can be specifically based on aspects such as the parameter's numerical range, the rationality of the state, and the correlation between the parameter and the state. For example, in a motor control scenario, if the current parameter of the input signal is 10A, and the system's preset normal operating current range is 8-15A, and the operating state is high-speed operation, and the current value logically matches the high-speed operation state, then the parameter identification result can be regarded as a target parameter identification result; if the current value exceeds the normal range or is inconsistent with the operating state, it will not be regarded as a target parameter identification result. If it is judged to be a target parameter identification result, it will be regarded as a control partition of the input signal, which represents the control requirement area corresponding to the current input signal.

[0040] When constructing the mapping relationship between signal partitions and control partitions, the high-frequency feature area, ambiguous feature resolution area, and correlation feature association area within the signal partition are used as the basis, combined with the descriptive information and categories of the control partition. Information about the high-frequency feature area of ​​a signal partition includes the specific type and feature value range of the high-frequency feature within the area. For example, the high-frequency feature area of ​​a voltage signal may include information such as "voltage fluctuation frequency 100Hz±5Hz" and "voltage amplitude 220V±10V." Information about the ambiguous feature resolution area includes the range of the resolution change rate of the ambiguous feature and the possible resolution directions. For example, an ambiguous feature resolution change rate between 0.5 and 1.2 may correspond to different control logic branches. Information about the correlation feature association area includes the type of correlation feature and the correlation strength value. For example, the correlation strength between temperature and pressure is 0.8 (maximum value 1.0).

[0041] The descriptive information of a control partition is a textual description of the partition's function, such as "motor start control partition" or "temperature adjustment control partition," clarifying the specific role of the partition in system control. The categories of control partitions are divided according to the control object or control method, such as "voltage control class," "speed control class," and "safety protection class." When establishing a mapping relationship, a feature matching algorithm is used to compare the feature information of the signal partition with the descriptive information and category of the control partition. For example, when the high-frequency feature area of ​​the signal partition contains the current mutation characteristics during motor startup, and the descriptive information of the control partition is "motor start control partition" and the category is "motor control class," the system uses the feature matching algorithm to confirm the corresponding relationship between the two, thereby establishing a mapping between the signal partition and the control partition.

[0042] The construction of mapping relationships must ensure logical consistency and accuracy. For example, the resolution results of a feature in the ambiguous feature resolution area of ​​the signal partition must strictly correspond to the control logic branch of the control partition, and the feature correlation strength in the correlation feature correlation area must meet the control partition's requirements for multi-parameter coordinated control. This mapping relationship is stored in the form of a data structure, such as a hash table or relational database table, where the key-value pairs are the signal partition feature identifier and the control partition identifier, respectively, to facilitate subsequent call queries by parameter combination modules, state update modules, etc.

[0043] Example 3: The implementation process of the parameter combination module revolves around the feature extraction and parameter combination analysis of signal partitions and control partitions, aiming to extract key partitions from complex signal and control information, and determine the combination matching degree between parameters through feature similarity analysis, and finally obtain parameter bias combination. The module first performs cluster analysis on the signal partitions and control partitions. The clustering dimensions include control type, function type and execution function. Among them, the control type can be divided into voltage control, current control, temperature control, etc.; the function type covers startup, speed regulation, protection, shutdown, etc.; the execution function involves the execution of specific actions, such as relay closure, valve adjustment, motor speed adjustment, etc.

[0044] Cluster analysis uses appropriate algorithms, such as the K-means algorithm or a hierarchical clustering algorithm based on distance metrics. The clustering process uses the feature vectors of the signal and control partitions as input. These feature vectors contain information describing the partitions, their feature types, and their eigenvalue ranges. By calculating the distances (such as Euclidean or Manhattan distances) between the feature vectors of each partition, partitions with high feature similarity are grouped together to form multiple clusters. After clustering is complete, the cluster center with the largest number of partitions in each cluster is designated as the key partition. This key partition represents the primary characteristics and control requirements of the cluster.

[0045] When extracting key features for critical partitions, it's necessary to analyze their composition and identify those features that are crucial for control decisions. For example, in a critical partition for temperature control, key features might include temperature measurements, temperature change rates, and temperature control targets. In a critical partition for motor speed regulation, key features might include current, voltage, and motor speed feedback. Extracting these key features requires integrating the device's control logic and process requirements to ensure that the extracted features accurately reflect the control nature of the critical partition.

[0046] When calculating feature similarity between key features, use an appropriate similarity metric, such as cosine similarity or the correlation coefficient. Cosine similarity measures similarity by calculating the cosine of the angle between two feature vectors, with values ​​closer to 1 indicating higher similarity. The correlation coefficient measures the degree of linear correlation between features. For example, for the two key features of temperature measurement and temperature control target, calculate the cosine similarity between their feature vectors to determine the degree of match.

[0047] When setting a common sequence based on the feature similarity between key features, we first identify key feature combinations with high similarity. These combinations often have synergistic effects during the control process. The common sequence is an ordered arrangement of these feature combinations, reflecting the logical order and relationships between features. For example, in the key partition of motor start control, key features include the start button signal, initial motor speed, and starting current threshold. If the start button signal and starting current threshold have high feature similarity, they are included in the common sequence and arranged according to the logical order of start control.

[0048] Utilizing common sequences related to feature similarity between key features, parameters within these common sequences are extracted. These parameters are specific numerical values ​​or state descriptions of key features, such as the starting current threshold value or the motor's initial speed level. When determining the longest common subsequence between parameters, string matching or dynamic programming algorithms are used to identify the longest continuous or non-continuous common subsequence within the parameter sequence. The length of the longest common subsequence reflects the degree of matching between the parameters; a longer length indicates a higher degree of matching between the parameter combinations.

[0049] After setting the length of the longest common subsequence as the combination matching degree between the parameters, the parameter bias combination needs to be set according to the time distribution probability of each parameter. The time distribution probability of each parameter is obtained through statistical analysis of historical operating data and reflects the probability of the parameter occurring at different time points. For example, by analyzing historical data of the motor startup process, it is found that the probability of the starting current threshold occurring at the startup moment is 0.8, and the probability of the motor initial speed occurring within 5 seconds after startup is 0.9.

[0050] When setting parameter bias combinations, the combination matching degree is combined with the temporal distribution probability. Parameter combinations with a high matching degree and a high temporal distribution probability are assigned a higher bias weight, giving them priority in the parameter bias combination. Parameter combinations with a low matching degree or a low temporal distribution probability are assigned a lower bias weight. This approach creates a parameter bias combination that takes into account both the matching degree between parameters and the temporal patterns of their occurrence, better meeting the actual operational requirements of the device.

[0051] For example, in a temperature control scenario, the key features of the key partition include the current temperature measurement, the temperature control target, and the heating element power. Calculations show a high degree of feature similarity between the current temperature measurement and the temperature control target, and their common sequence contains the temperature difference parameter and the heating time parameter. After extracting these two parameters, the length of their longest common subsequence is calculated to be 3, indicating a high degree of matching between the corresponding combinations. Furthermore, historical data statistics show that the temperature difference parameter has a probability of 0.7 appearing in the early stages of temperature control, and the heating time parameter has a probability of 0.8 appearing during the control process. Therefore, when setting the parameter bias combination, the combination of the temperature difference parameter and the heating time parameter is prioritized to guide subsequent temperature control strategies.

[0052] Example 4: The implementation of the state update module is based on the parameter bias combination. By analyzing the parameter time distribution probability and setting and fitting the target path, the control partition state is updated. Specifically, the module first extracts the time distribution probability of each parameter from the parameter bias combination. These probability values ​​are derived from the statistical analysis of historical operating data and reflect the probability of the parameters appearing in different time periods. For example, in a motor speed control system, the parameter bias combination may include the current threshold, the voltage regulation amount, and the speed feedback value. The time distribution probability of the current threshold in the startup phase (0-5 seconds) is 0.9, the probability of the voltage regulation amount in the speed control phase (5-15 seconds) is 0.8, and the probability of the speed feedback value in the stable operation phase (after 15 seconds) is 0.7.

[0053] After extracting the time distribution probability, the parameter bias combination is set as the target path according to the time period corresponding to each parameter. The target path is an orderly planning of the parameters in the time dimension, reflecting the expected change trajectory of the parameter combination at different stages. Taking the motor speed regulation scenario as an example, the target path in the startup phase is that the current threshold rises rapidly from the initial value to the set value, the target path in the speed regulation phase is that the voltage regulation amount is gradually adjusted according to the speed feedback value, and the target path in the stable operation phase is that the speed feedback value remains within the target range. The division of each time period is determined by the operating process and control logic of the equipment. For example, the duration of the startup phase, speed regulation phase, and stabilization phase can be set according to the motor type and load conditions.

[0054] When fitting the target path for each parameter in the parameter bias combination, an appropriate fitting method, such as polynomial fitting or spline curve fitting, is used to obtain a continuous target path curve. During the fitting process, the temporal distribution probability of each parameter is taken into account to ensure that the fitted target path better matches the parameter's changing trend during actual operation. For example, the current threshold has a higher temporal distribution probability during the startup phase, so the fitting process emphasizes the steepness of the current rise during this phase to match the high-probability requirement for rapid startup. The voltage regulation amount has a higher probability during the speed regulation phase, so the fitting process smoothes the slope of the voltage adjustment to match the probability distribution characteristics of this phase.

[0055] The probability value of the fitted target path in each time period is set as the probability of occurrence of the parameter bias combination. This probability value comprehensively reflects the likelihood of a parameter combination occurring within a specific time period. For example, during the motor startup phase (0-5 seconds), the probability of occurrence of the combination of current threshold, voltage regulation, and speed feedback is 0.85, indicating that this parameter combination has a high probability of occurrence during the startup phase. The calculation of the probability of occurrence of the combination is based on the joint probability of the time distribution probabilities of each parameter, taking into account the correlation and temporal order of the parameters.

[0056] Next, the probability of occurrence of the parameter bias combinations is compared with the state references in the control partition. The state references in the control partition record the frequency and time points of each state call during actual operation. For example, in the control partition of a motor speed control system, the "startup state" is referenced 100 times from 0 to 5 seconds, the "speed regulation state" is referenced 150 times from 5 to 15 seconds, and the "stable state" is referenced 200 times after 15 seconds. During the comparison, the probability of occurrence of the combination is matched with the time distribution of the state references to identify the difference between the two.

[0057] For example, the probability of a parameter bias combination occurring in the startup phase is 0.85, while the proportion of citations of the "startup state" in the control partition during this phase is 0.75, resulting in a difference of 0.1. The difference value can be calculated using either an absolute difference or a relative difference, depending on the system requirements. After identifying all the difference values ​​that appear, the states in the control partition are classified according to the size and direction of the difference values. The classification criteria can be divided into "high difference state", "medium difference state" and "low difference state", where a high difference state refers to a state where the difference value exceeds the preset threshold and needs to be updated first.

[0058] After completing the state classification, the states in the control partition are updated. The update method is determined by the nature of the difference. For example, if the probability of a combination occurring is higher than the number of state references, it means that the state of the control partition does not fully reflect the actual needs of the parameter combination, and the relevant state needs to be added or strengthened. If the probability of a combination occurring is lower than the number of state references, it means that the control partition has redundant states and needs to be optimized or deleted. Taking the motor speed regulation scenario as an example, if the probability of a combination occurring in the startup phase is 0.85, and the number of references to the "startup state" is 0.75, it is a high-difference state. A "quick start sub-state" can be added to the control partition to match the high-probability needs of the parameter combination.

[0059] During the entire state update process, taking the motor temperature control scenario as an example, the parameter bias combination may include the temperature setpoint, heating power, and cooling fan speed. The time distribution probability of each parameter is as follows: the temperature setpoint has a probability of 0.9 during the initial warm-up phase (0-30 minutes), the heating power has a probability of 0.8 during this phase, and the cooling fan speed has a probability of 0.7 during the temperature stabilization phase (after 30 minutes). When setting the target path, the target path for the initial warm-up phase is to gradually increase the temperature setpoint, maintain the maximum heating power, and minimize the cooling fan speed. The target path for the temperature stabilization phase is to maintain the temperature setpoint unchanged, gradually reduce the heating power, and adjust the cooling fan speed based on temperature fluctuations.

[0060] After fitting the target path, the probability of the combination occurring during the initial heating phase is 0.88, while the reference count for the "heating state" in the control partition is 0.80, with a variance of 0.08. Since the variance does not exceed the preset threshold of 0.1, this represents a medium variance state. Fine-tune the control logic for the "heating state," such as shortening the duration of the maximum heating power, to match the probability of the parameter combination. The probability of the combination occurring during the temperature stabilization phase is 0.75, while the reference count for the "stable state" in the control partition is 0.90, with a variance of -0.15. This represents a high variance state. A new "heating optimization sub-state" should be added to the control partition to reflect the probability distribution characteristics of the cooling fan speed parameter.

[0061] Through the aforementioned process, the state update module dynamically adjusts the state of control partitions, ensuring that they accurately reflect the actual needs of parameter-biased combinations. From extracting the temporal distribution probability to setting the target path, to fitting and calculating the probability of occurrence of the combination, and finally completing the state update through difference comparison and classification, each step closely revolves around the temporal correlation between parameters and states. This enables the intelligent electronic control system to optimize the state of control partitions in real time based on the probabilistic distribution characteristics of parameter combinations, improving the system's adaptability and control accuracy. This process does not rely on hypothetical experimental data, but rather on actual parameter distribution and state references. It achieves reasonable state updates through logical analysis and data processing, providing accurate state information support for subsequent control management modules.

[0062] Example 5: The implementation process of the control management module is based on the updated mapping relationship between signal partitions and control partitions. By extracting control objectives, determining control sequences, and generating a control management library, it achieves systematic management of input signals. For example, in an industrial boiler temperature control system, when the system detects the input signal from the boiler water temperature sensor, the signal partitions have been previously divided into high-frequency feature areas (such as water temperature change rate characteristics), ambiguous feature resolution areas (such as sensor signal fluctuation ambiguity), and correlation feature association areas (such as the correlation between water temperature and gas flow). The control partitions correspond to different functional areas such as the "temperature rise control area" and the "constant temperature control area." At this point, the control management module begins processing.

[0063] The module extracts the control objective from the input signal based on the updated mapping relationship. In a boiler temperature scenario, if the current water temperature is 60°C and the preset target temperature is 80°C, and the high-frequency feature area of ​​the signal partition indicates that the water temperature is increasing at a rate of 2°C / minute, the control objective of "raising the water temperature to 80°C" can be extracted in combination with the mapping relationship of the control partition. The extraction of the control objective requires combining the feature analysis results of the signal partition with the functional definition of the control partition. For example, if the correlation feature association area of ​​the signal partition indicates a high correlation between water temperature and gas valve opening, the control objective may involve adjusting the gas valve opening.

[0064] After extracting the control objectives, the module sorts them by their probability of occurrence to determine the control sequence. The probability of occurrence of control objectives is determined through historical data statistics. For example, in routine boiler operation, the probability of occurrence of the "temperature increase control objective" during the startup phase is 0.9, and the probability of occurrence of the "constant temperature control objective" during the stable operation phase is 0.8. Taking the startup phase as an example, the control objectives of the input signal may include "opening the gas valve," "starting the circulating water pump," and "monitoring the water temperature." Based on historical statistics, the probability of "opening the gas valve" is 0.95, the probability of "starting the circulating water pump" is 0.9, and the probability of "monitoring the water temperature" is 0.85. Therefore, the control sequence is determined to execute "opening the gas valve" first, "starting the circulating water pump" second, and "monitoring the water temperature" last.

[0065] The logical dependencies between control objectives must be considered during the sequencing process. For example, "monitoring water temperature" must be executed after "starting the circulating water pump," even if its probability of occurrence is low. In boiler temperature control, if the probability of "adjusting gas flow" is 0.8, while the probability of "monitoring water temperature" is 0.7, but "adjusting gas flow" depends on the result of "monitoring water temperature," the control sequence should be adjusted to prioritize "monitoring water temperature" over "adjusting gas flow" to ensure the rationality of the control logic.

[0066] After sorting the control objectives, the module combines the control objectives and control sequence in a structured format to generate a control management library. This structured format can take the form of a table, tree structure, or database record. Taking the table format as an example, the control management library may contain fields such as "Control Target Number," "Control Target Description," "Execution Order," "Associated Signal Partition," and "Corresponding Control Partition." In a boiler temperature control scenario, a record in the control management library might be: Number 001, Control Target "Open Gas Valve," Execution Order 1, Gas Valve Open Signal Feature Zone in the Associated Signal Zone, and "Temperature Rise Control Zone" in the corresponding control zone.

[0067] The generation of the control management library must ensure information integrity and traceability. Each control objective must clearly identify the associated signal partition characteristics and corresponding control partition to facilitate subsequent querying and debugging. For example, if the "regulate gas flow" control objective performs abnormally, the control management library can quickly locate the associated signal partition (such as the high-frequency characteristic area of ​​the gas flow sensor) and control partition (constant temperature control area), facilitating troubleshooting.

[0068] For a control circuit board with this intelligent electronic control system, its hardware implementation is a hardware circuit unit that integrates a signal acquisition module, a pattern mapping module, a parameter combination module, a state update module, and a control management module. Taking a boiler control circuit board as an example, the hardware circuit of the signal acquisition module may include interfaces for water temperature sensors and gas flow sensors, etc., for receiving input signals; the hardware circuit of the pattern mapping module includes a microprocessor chip for performing parameter identification and mapping relationship construction; the hardware circuit of the parameter combination module may include a memory chip for storing historical data and cluster analysis results; the hardware circuit of the state update module includes timers and logic gate circuits for processing time distribution probabilities and state update logic; and the hardware circuit of the control management module includes a communication interface and a storage unit for generating and storing the control management library.

[0069] Each hardware circuit unit realizes signal transmission and data interaction through circuit connection. For example, the water temperature signal obtained by the signal acquisition module is transmitted to the microprocessor of the mode mapping module through the copper foil line on the circuit board. After processing, the microprocessor transmits the mapping relationship data to the storage chip of the parameter combination module through the bus. The parameter bias combination obtained by the parameter combination module is then transmitted to the timer and logic gate circuit of the status update module through the data line. Finally, the control management module obtains the processing results from each module, generates a control management library, and outputs it to the external control device through the communication interface.

[0070] In the specific workflow, when the boiler is started, the water temperature sensor input signal is processed by the signal acquisition module and divided into different signal partitions. The pattern mapping module maps it to the "temperature rise control zone". The parameter combination module extracts key features from the signal partition and the control partition (such as the initial water temperature value, target temperature, and gas flow threshold), and calculates the parameter bias combination (such as prioritizing the adjustment of the gas valve opening to 50% and maintaining this state until the water temperature reaches 70°C). The state update module updates the state of the control partition (such as adding a "rapid temperature rise state") according to the time distribution probability of the parameter (such as the probability of the gas valve opening being 0.9 0-10 minutes after startup). The control management module finally extracts the control targets (opening the gas valve, adjusting the valve opening, and monitoring the water temperature), sorts them according to the probability of occurrence and logical order, and generates a control management library. The control circuit board executes the instructions in the control management library through the hardware circuit to achieve precise control of the boiler temperature.

[0071] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0072] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent electronic control system, characterized in that: include: The signal acquisition module is used to obtain the input signal when the equipment is running, and determine the signal partition of the input signal relative to the electronic control function according to the control requirements of the equipment during operation; A pattern mapping module is used to obtain the control partition of the input signal and build a mapping relationship between the signal partition and the control partition; The parameter combination module is used to extract key partitions from the signal partition and the control partition, and judge the combination matching degree between multiple parameters in the control partition according to the key partition to obtain the parameter bias combination; A state update module is used to verify the information of the control partition based on the parameter bias combination, identify the update of the state in the control partition, and update the mapping relationship between the signal partition and the control partition according to the update of the state; The control management module is used to identify the control target and control sequence of the input signal according to the updated mapping relationship between the signal partition and the control partition, and generate a control management library.

2. The intelligent electronic control system according to claim 1, characterized in that: The implementation of the signal acquisition module includes: For any input signal during device operation, obtain a type recognition model corresponding to the input signal; Classify the input signal using a type recognition model to obtain at least one type category; Identify high-frequency features, ambiguous features, and related features in the input signal under the corresponding type category to form a signal feature set; The high-frequency features, ambiguous features and associated features in the signal feature set are analyzed respectively, and the high-frequency feature area, ambiguous feature resolution area and associated feature association area corresponding to the signal feature set are obtained in turn, and used as the signal partitioning of the input signal relative to the electronic control function.

3. The intelligent electronic control system according to claim 2, characterized in that: The implementation method of obtaining the high-frequency feature area, the ambiguous feature resolution area, and the associated feature association area corresponding to the signal feature set also includes: Merge the high-frequency features, ambiguous features, and associated features in the signal feature set according to type categories to obtain multiple type merging results; Extract high-frequency feature pairs from the type merging results, compare the high-frequency feature pairs with the feature library, and obtain high-frequency feature areas; The resolution change rate of the ambiguous features and the association strength of the associated features in the type merging result are extracted, and the type merging result is divided according to the resolution change rate of the ambiguous features and the association strength of the associated features to obtain the ambiguous feature resolution area and the associated feature association area.

4. The intelligent electronic control system according to claim 1, characterized in that: The implementation method of determining the signal partition of the input signal relative to the electronic control function also includes: Judge the signal partitions, analyze the signal times and signal times of the devices in the signal partitions, fit the signal partitions according to the signal times and signal times, and build a mapping relationship between the signal partitions and the actual needs of the devices.

5. The intelligent electronic control system according to claim 1, characterized in that: The implementation of the pattern mapping module includes: Call the parameters and states corresponding to the input signal to generate multiple unlabeled parameter recognition results. The unlabeled parameter recognition results represent the parameters and states that are not associated with the features in the signal partition; It is determined whether the multiple unlabeled parameter recognition results are target parameter recognition results. If they are target parameter recognition results, the target parameter recognition results are regarded as the control partitions of the input signal.

6. The intelligent electronic control system according to claim 3, characterized in that: The implementation method of constructing the mapping relationship between the signal partition and the control partition includes: using the information representation of the high-frequency feature area, ambiguous feature resolution area and associated feature association area existing in the signal partition, the descriptive information of the control partition and the category of the control partition to construct a mapping relationship between the signal partition and the control partition.

7. The intelligent electronic control system according to claim 1, characterized in that: The implementation of the parameter combination module includes: Perform cluster analysis on signal partitions and control partitions according to control type, function type, and execution function, and set the largest cluster center after cluster analysis as the key partition; Extracting key features of key partitions, calculating feature similarities between key features, and setting a common sequence related to the feature similarities between key features; Using the common sequence related to the feature similarity between the key features, the parameters existing in the common sequence are extracted, and the longest common subsequence between the parameters is set, and the length value of the longest common subsequence is set as the combined matching degree between the parameters; The combination matching degree between the parameters is set according to the time distribution probability of each parameter, and the parameter bias combination is set.

8. The intelligent electric control system according to claim 7, characterized in that: The implementation of the status update module includes: Extracting the time distribution probability of each parameter from the parameter bias combination; setting the target path of the parameter bias combination according to the time period corresponding to the time distribution probability of each parameter; Fitting the target path of each parameter in the parameter bias combination to obtain a fitted target path, and setting the probability value of the fitted target path in each time period as the combined occurrence probability of the parameter bias combination; The probability of occurrence of the parameter bias combination is compared with the reference of the state in the control partition, the difference value is identified, and the state in the control partition is classified according to the difference value, thereby completing the update of the state in the control partition.

9. The intelligent electronic control system according to claim 1, characterized in that: The control management module is implemented as follows: According to the mapping relationship between the updated signal partition and the control partition, the control targets in the input signal are extracted, and the control targets are sorted according to the probability of occurrence of the control targets to obtain the control sequence in the input signal; the control targets and the control sequence are combined in a structured form to obtain a control management library.

10. A control circuit board of the intelligent electronic control system according to any one of claims 1 to 9, characterized in that: include: A hardware circuit unit integrating the signal acquisition module, mode mapping module, parameter combination module, state update module and control management module realizes signal transmission and data interaction between modules through circuit connection.

Images