Emergency stop control method for mechanical and electrical equipment based on real-time monitoring

Abstract

The invention belongs to the technical field of control, and discloses a mechanical electrical equipment sudden stop control method based on real-time monitoring. Comprising the steps of obtaining a mechanical electrical equipment operation state feature vector through mechanical electrical equipment state parameters; carrying out weighted fusion through each parameter and the corresponding dynamic reference adaptive membership degree value to obtain a risk situation entropy value; selecting a communication toughness strategy based on the risk situation entropy, and adjusting a heartbeat packet sending frequency to construct a communication channel robustness index; according to the risk situation entropy, the channel robustness index and the mechanical and electrical equipment state parameters, calculating emergency shutdown confidence; constructing a comprehensive risk triggering probability based on the emergency shutdown confidence coefficient and the channel robustness index; setting a multi-factor decision based on the risk trigger probability; using the risk situation entropy, the channel robustness index and the emergency shutdown confidence to calculate a compensation effect index to trigger a risk decision or a multi-factor decision; and the safety guarantee capability is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of control technology, specifically to an emergency stop control method for mechanical and electrical equipment based on real-time monitoring. Background Technology

[0002] Existing emergency stop control methods for mechanical and electrical equipment mostly rely on fixed thresholds or periodic detection, lacking dynamic tracking of the equipment's status. Furthermore, regardless of the data's state, they only employ a single transmission mode, failing to guarantee real-time transmission of critical commands. Fixed-frequency heartbeat protocols are also inadequate to handle sudden changes in risk. Moreover, existing decision-making logic is overly simplistic, typically relying on a single parameter or condition combination to trigger an emergency stop, neglecting comprehensive multi-dimensional risk assessment, leading to inaccurate or delayed decisions. Additionally, existing emergency stop control methods for mechanical and electrical equipment generally cease at the triggering stage, lacking closed-loop control verification, which easily leads to control failure.

[0003] In view of this, the present invention proposes an emergency stop control method for mechanical and electrical equipment based on real-time monitoring to solve the above problems. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: an emergency stop control method for mechanical and electrical equipment based on real-time monitoring, comprising:

[0005] Step SS1: Drive the heterogeneous sensor array to synchronously acquire the state parameters of mechanical and electrical equipment in the time domain; analyze the state parameters of mechanical and electrical equipment using sliding time window analysis to establish the operating state feature vector of mechanical and electrical equipment; backtrack to extract the operating state feature vector of mechanical and electrical equipment from N historical time steps and input it into the LSTM model to generate adaptive confidence boundaries; use the dynamic benchmark adaptive membership function to calculate the dynamic benchmark adaptive membership value of each parameter in the state parameters of mechanical and electrical equipment; perform multi-source feature tensor fusion based on each parameter and its corresponding dynamic benchmark adaptive membership value to obtain the risk situation entropy value;

[0006] Step SS2: Dynamically select a communication resilience strategy using the risk situation entropy value in the risk selection rules. Under the selected communication resilience strategy, adopt an adaptive heartbeat synchronization strategy to adjust the heartbeat packet transmission frequency. Calculate the packet loss rate based on the heartbeat packet transmission frequency, and construct a channel robustness index based on the packet loss rate and transmission jitter.

[0007] Step SS3: Make risk decisions based on the risk situation entropy value, channel robustness index, and mechanical and electrical equipment status parameters, and calculate the emergency stop confidence level; construct a comprehensive risk trigger probability based on the emergency stop confidence level and the channel robustness index; set multi-factor decisions based on the risk trigger probability.

[0008] Step SS5: Compensate for the risk situation entropy value, channel robustness index, and emergency shutdown confidence level; calculate the compensation effect index based on the compensated risk situation entropy value, compensated channel robustness index, compensated emergency shutdown confidence level, and mechanical and electrical equipment operating status feature vector; trigger risk decision-making or multi-factor decision-making based on the compensation effect index.

[0009] Furthermore, the specific methods for obtaining the risk situation entropy value include:

[0010] The status parameters of mechanical and electrical equipment are acquired synchronously in the time domain through a heterogeneous sensor array.

[0011] Sliding time window analysis is used to analyze the state parameters of mechanical and electrical equipment, and to establish the operating state feature vector of the mechanical and electrical equipment; wherein, the window length T and the sliding step size are set. This ensures continuous coverage and real-time performance of the data stream; it calculates statistical quantities and frequency domain characteristics of the mechanical and electrical equipment status parameters within each time window. The statistical quantities include mean, variance, skewness, and kurtosis, while the frequency domain characteristics include low-frequency, mid-frequency, and high-frequency quantities.

[0012] Obtain the feature vectors of the operating state of mechanical and electrical equipment over N historical time steps, and input them into the LSTM model to generate adaptive confidence boundaries.

[0013] The state parameters of mechanical and electrical equipment are normalized, and the dynamic benchmark adaptive membership function is used to calculate the dynamic benchmark adaptive membership value of each parameter.

[0014] Risk situation entropy is obtained by multi-source feature tensor fusion based on each parameter and its corresponding dynamic benchmark adaptive membership value.

[0015] Furthermore, the specific method for obtaining the dynamic benchmark adaptive membership degree value of each parameter in the calculation of the state parameters of mechanical and electrical equipment using the dynamic benchmark adaptive membership function includes:

[0016] Use linear functions, parabolic functions, and S-curve functions as reference functions;

[0017] The benchmark function is dynamically selected based on the statistics of each parameter; the statistics include standard deviation, skewness, and kurtosis.

[0018] Select the corresponding benchmark function based on the statistic. If the standard deviation is within the standard range, choose the linear function and calculate the membership value; if the skewness is within the standard range, choose the parabolic function and calculate the membership value; if the kurtosis is within the standard range, choose the S-curve and calculate the membership value.

[0019] If the standard deviation is greater than the standard range, select a membership value of 1; if the skewness is greater than the standard range, select a membership value of 1; if the kurtosis is greater than the standard range, select a membership value of 1.

[0020] If the standard deviation is less than the standard range, select a membership value of 0; if the skewness is less than the standard range, select a membership value of 0; if the kurtosis is less than the standard range, select a membership value of 0.

[0021] The weights of the statistics for each parameter are optimized using gradient descent, and then weighted fusion is performed to obtain the dynamic baseline adaptive membership value.

[0022] Furthermore, the specific method for dynamically selecting a communication resilience strategy using risk situation entropy values ​​in the risk selection rules, and adjusting the heartbeat packet transmission frequency using an adaptive heartbeat synchronization strategy under the selected communication resilience strategy, includes:

[0023] The risk selection rules include: when the risk situation entropy value is within the adaptive confidence boundary, dual-channel redundancy verification is activated; when the risk situation entropy value is greater than the adaptive confidence boundary, three-channel mirror transmission is activated; when the risk situation entropy value is less than the adaptive confidence boundary, single-channel economic mode is adopted.

[0024] Dual-channel redundancy verification, three-channel mirror transmission, and single-channel economic mode are communication resilience strategies;

[0025] Retrieve risk situation entropy values ​​over F historical time periods and calculate their moving average; set a baseline frequency based on the moving average; the formula for the baseline frequency is: Baseline Frequency = *Sliding average+ ,in, and To adjust the parameters and ensure that the reference frequency remains within a reasonable range;

[0026] The initial transmission frequency is constructed by mapping the current risk situation entropy value to the base frequency; if the current risk situation entropy value is less than the adaptive confidence boundary, the initial transmission frequency is 50% of the base frequency to reduce communication resource consumption.

[0027] If the current risk situation entropy value is within the adaptive confidence boundary, the initial transmission frequency is an exponential increase of the base frequency, as shown in the formula: ,in, This is the initial transmission frequency. Indicates the reference frequency. Indicates response sensitivity. Control the steepness of the curve, This represents the risk situation entropy value;

[0028] If the current risk situation entropy value is greater than the adaptive confidence boundary, the initial transmission frequency is 150% of the base frequency;

[0029] The rate of change of the current risk situation entropy is introduced to dynamically correct the initial transmission frequency, thus obtaining the transmission frequency.

[0030] Furthermore, the specific methods for calculating the packet loss rate based on the heartbeat packet transmission frequency and constructing the channel robustness index based on the packet loss rate and transmission jitter include:

[0031] The packet loss rate is calculated by counting the total number of heartbeat packets sent within a time period and the number of unacknowledged responses received; the total number of heartbeat packets is the product of the time period and the heartbeat packet sending frequency.

[0032] The method for obtaining the number of heartbeat packets that did not receive an acknowledgment response is as follows: the sending frequency and sending time of the heartbeat packets are known; after each heartbeat packet is sent, it waits for an acknowledgment response (including acknowledgment number and sending timestamp) from the other end (such as the server). If no acknowledgment response is received within the set timeout period (such as 3 times RTT, i.e., round-trip time), it is determined that the heartbeat packet was not successfully received. The total number of heartbeat packets that did not receive an acknowledgment response during the accumulated time period is the number of heartbeat packets that did not receive an acknowledgment response.

[0033] Transmission jitter is defined as the standard deviation of the arrival time of heartbeat packets within a time period.

[0034] The channel robustness index is calculated based on packet loss rate and transmission jitter, using the following formula: , For channel robustness index, For packet loss rate, For heartbeat packet transmission jitter, For maximum transmission jitter, and These are the weights for packet loss rate and packet skipping jitter, respectively.

[0035] Furthermore, the risk decision-making based on the risk situation entropy value, channel robustness index, and mechanical and electrical equipment state parameters, and the calculation of the emergency shutdown confidence level; the specific methods for constructing the comprehensive risk trigger probability based on the emergency shutdown confidence level and the channel robustness index include:

[0036] Step aa1, Risk Decision: Input Nodes: The risk situation entropy value, channel robustness index, and mechanical and electrical equipment status parameters are used as input nodes;

[0037] Output node: The output node represents the emergency stop confidence level;

[0038] Step aa2: Set the comprehensive risk trigger probability of mechanical and electrical equipment based on the emergency stop confidence level and the channel robustness index;

[0039] The failure risk of the channel robustness index is calculated by comparing it with the theoretical maximum index of channel robustness; the failure rate of the channel robustness index is then calculated using the failure risk of the channel robustness index.

[0040] The overall risk trigger probability is calculated by weighting the emergency shutdown confidence level and the channel failure rate.

[0041] Furthermore, the specific methods for setting multi-factor decision-making based on risk trigger probability include:

[0042] Level 1 Emergency Stop: When the overall risk trigger probability is higher than 0.7, it indicates that the mechanical and electrical equipment is in an extremely high-risk state and an emergency stop operation will be performed. At this time, the system will send an emergency stop command to cut off the power to the entire system and lock the mechanical devices to ensure equipment safety.

[0043] Level 2 slow stop: When the overall risk trigger probability is less than 0.7 and greater than 0.5, it indicates that the mechanical and electrical equipment is in a high-risk state, but it is not necessary to immediately execute an emergency stop. At this time, the system will start a phased speed reduction (such as reducing the equipment running speed) and a partial power cut-off to reduce the equipment load and ensure a safe transition to the shutdown state.

[0044] Manual review mode: When the emergency shutdown confidence level is lower than 0.5 and the risk situation entropy value exceeds the adaptive confidence boundary E times consecutively within H historical time steps, it indicates that the current state is abnormal and manual intervention is required. At this time, an early warning is issued, and the system enters the manual review mode, requiring operators to perform manual review and judgment.

[0045] Furthermore, the specific methods for compensating for the risk situation entropy value, channel robustness index, and emergency shutdown confidence include:

[0046] Active dampers, passive dampers, and failure dampers are installed in mechanical and electrical equipment. The process damping coefficient is calculated by collecting the active damping coefficient, passive damping coefficient, and failure damping coefficient from the active dampers, passive dampers, and failure dampers. The process damping coefficient = a * active damping coefficient + b * passive damping coefficient + c * failure damping coefficient, where a, b, and c represent a active dampers, b passive dampers, and c failure dampers installed in the mechanical and electrical equipment, respectively.

[0047] For the risk situation entropy value, the process damping coefficient is used for compensation to obtain the compensated risk situation entropy value. The formula is: Compensated risk situation entropy value = Process damping coefficient * Risk situation entropy value;

[0048] Zigbee modules are deployed in mechanical and electrical equipment. The RSSI values ​​at L locations are recorded using a mobile receiver. The average RSSI value is calculated based on the RSSI values ​​at L locations. The average RSSI value is then used to compensate for the channel robustness index, resulting in a compensated channel robustness index.

[0049] To compensate for the emergency shutdown confidence level, the risk situation entropy value and the channel robustness index are used to obtain the compensated emergency shutdown confidence level.

[0050] Furthermore, the specific methods for calculating the compensation effect index based on the compensation risk situation entropy value, compensation channel robustness index, compensation emergency stop confidence level, and mechanical and electrical equipment operating state feature vector include:

[0051] The weights for compensating for risk situation entropy are calculated using the exponential decay method; the weights for compensating for channel robustness index are calculated using the sigmoid function method; and the weights for compensating for emergency shutdown confidence are calculated using the rate of change method.

[0052] The comprehensive contribution value is obtained by multiplying the compensation risk situation entropy value, the compensation channel robustness index, and the compensation emergency shutdown confidence value and their corresponding weights.

[0053] The total weight value is obtained by adding the weights of the compensation risk situation entropy value, the compensation channel robustness index, and the compensation emergency shutdown confidence value.

[0054] The overall evaluation value is obtained by comparing the comprehensive contribution value and the total weight value.

[0055] The deviation ratio is calculated by comparing the ratio of the mechanical and electrical equipment condition parameters to the mechanical and electrical equipment condition difference parameters; where the mechanical and electrical equipment condition difference parameters are the differences between the mechanical and electrical equipment condition parameters and the mechanical and electrical equipment condition standard parameters.

[0056] The compensation effect index is calculated using the total evaluation value and the deviation ratio.

[0057] Furthermore, the specific methods for triggering risk decisions or multi-factor decisions based on the compensation effect index include:

[0058] If the compensation effect index is greater than If the compensation effect is unsatisfactory, a risk decision is triggered, the risk trigger probability is recalculated, and new factor trigger conditions are established based on the new risk trigger probability; among which, For channel robustness index, This is the adjustment coefficient;

[0059] If the compensation effect index is greater than J times consecutively This triggers a first-level emergency stop in multi-factor decision-making.

[0060] The technical effects and advantages of the emergency stop control method for mechanical and electrical equipment based on real-time monitoring in this invention are as follows:

[0061] This invention establishes a feature vector of the operating state of mechanical and electrical equipment through the state parameters of the equipment. The sliding time window analysis and multi-dimensional feature extraction (statistics / frequency domain features) ensure the timeliness of the data and solve the delay problem caused by traditional batch processing.

[0062] The adaptive confidence boundary is calculated using an LSTM model, and the dynamic baseline adaptive membership value of each mechanical and electrical equipment state parameter is calculated using a dynamic baseline membership function. The baseline function is intelligently selected based on the parameter statistical characteristics (standard deviation / skewness / kurtosis) to achieve adaptive adjustment of anomaly sensitivity.

[0063] The risk situation entropy value is not only based on the current state assessment, but also integrates a weighted model of dynamic membership function and multi-dimensional features, which effectively improves the accuracy and stability of the judgment.

[0064] By using an adaptive heartbeat synchronization strategy, the heartbeat frequency is dynamically adjusted, reducing the waste of communication resources while dynamically responding to the degree of danger, thus improving the system response efficiency. The channel robustness index not only considers the packet loss rate but also introduces transmission jitter, providing a more comprehensive assessment of data link stability.

[0065] Different levels of emergency stop strategies (Level 1 emergency stop, Level 2 gradual stop, and manual review) achieve "flexible shutdown" and avoid unnecessary economic losses caused by a one-size-fits-all emergency stop.

[0066] A comprehensive compensation method based on process damping compensation, communication RSSI compensation, and emergency shutdown confidence compensation can effectively repair system fluctuations caused by noise disturbances, false alarms, and missed alarms.

[0067] Ultimately, by using the compensation effect index, a closed-loop risk decision is triggered, enabling the system to have self-diagnosis, self-repair, and self-decision-making capabilities. Attached Figure Description

[0068] Figure 1 This is a schematic diagram of the emergency stop control method for mechanical and electrical equipment based on real-time monitoring according to the present invention;

[0069] Figure 2 This is a schematic diagram of the communication assurance subsystem of the present invention;

[0070] Figure 3 This is a schematic diagram of the emergency stop control system for mechanical and electrical equipment based on real-time monitoring, as described in this invention. Detailed Implementation

[0071] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0072] Example 1

[0073] Please see Figure 1 and Figure 2 As shown, this embodiment of the emergency stop control method for mechanical and electrical equipment based on real-time monitoring includes:

[0074] Step SS1: Drive the heterogeneous sensor array to synchronously acquire the state parameters of mechanical and electrical equipment in the time domain; analyze the state parameters of mechanical and electrical equipment using sliding time window analysis to establish the operating state feature vector of mechanical and electrical equipment; backtrack to extract the operating state feature vector of mechanical and electrical equipment from N historical time steps and input it into the LSTM model to generate adaptive confidence boundaries; use the dynamic benchmark adaptive membership function to calculate the dynamic benchmark adaptive membership value of each parameter in the state parameters of mechanical and electrical equipment; perform multi-source feature tensor fusion based on each parameter and its corresponding dynamic benchmark adaptive membership value to obtain the risk situation entropy value;

[0075] Step SS2: Dynamically select a communication resilience strategy using the risk situation entropy value in the risk selection rules. Under the selected communication resilience strategy, adopt an adaptive heartbeat synchronization strategy to adjust the heartbeat packet transmission frequency. Calculate the packet loss rate based on the heartbeat packet transmission frequency, and construct a channel robustness index based on the packet loss rate and transmission jitter.

[0076] Step SS3: Make risk decisions based on the risk situation entropy value, channel robustness index, and mechanical and electrical equipment status parameters, and calculate the emergency stop confidence level; construct a comprehensive risk trigger probability based on the emergency stop confidence level and the channel robustness index; set multi-factor decisions based on the risk trigger probability.

[0077] Step SS5: Compensate for the risk situation entropy value, channel robustness index, and emergency shutdown confidence level; calculate the compensation effect index based on the compensated risk situation entropy value, compensated channel robustness index, compensated emergency shutdown confidence level, and mechanical and electrical equipment operating status feature vector; trigger risk decision-making or multi-factor decision-making based on the compensation effect index.

[0078] The specific methods for obtaining risk situation entropy values ​​include:

[0079] Mechanical and electrical equipment status parameters (flow rate, speed, voltage, frequency, load current, ambient temperature, vibration amplitude, equipment load rate, grounding resistance, insulation resistance, bearing temperature, lubricating oil pressure, lubricating oil temperature, current harmonics, voltage harmonics) are acquired synchronously in the time domain through a heterogeneous sensor array.

[0080] Sliding time window analysis is used to analyze the state parameters of mechanical and electrical equipment, and to establish the operating state feature vector of the mechanical and electrical equipment; wherein, the window length T and the sliding step size are set. This ensures continuous coverage and real-time performance of the data stream; it calculates statistical quantities and frequency domain characteristics of the mechanical and electrical equipment status parameters within each time window. The statistical quantities include mean, variance, skewness, and kurtosis, while the frequency domain characteristics include low-frequency, mid-frequency, and high-frequency quantities.

[0081] For example: for a voltage sensor with a sampling rate of 10kHz (i.e., 10,000 data points per second), a window length of T=1 second, and a sliding step size... Milliseconds, or 0.1 seconds, are measured by sliding 1000 data points each time. The first window contains 0 to 1 second of time, the second window contains 0.1 to 1.1 seconds of time, and so on.

[0082] The voltage data sequence in window 1 is , This is the voltage data value of the first sampling point. The voltage data value at the 10,000th sampling point is used; the voltage statistics and frequency domain features within the time window are calculated; and the statistics and frequency domain features are fused into a voltage feature vector with a dimension of 7. The dimension of the mechanical and electrical equipment operating state feature vector is the data included in the mechanical and electrical equipment state parameters multiplied by 7, resulting in the total dimension of the mechanical and electrical equipment operating state feature vector. If the mechanical and electrical equipment state parameters include 16 parameter data, the total dimension of the mechanical and electrical equipment operating state feature vector is 16*7=112 dimensions.

[0083] Obtain the feature vectors of the operating state of mechanical and electrical equipment over N historical time steps, and input them into the LSTM model to generate adaptive confidence boundaries.

[0084] Example: Obtain the feature vector of the operating status of mechanical and electrical equipment over 1000 historical time steps. There are a total of 1000 sets of feature vectors for the operating status of mechanical and electrical equipment; the output is an adaptive confidence boundary. ;

[0085] The state parameters of mechanical and electrical equipment are normalized, and the dynamic benchmark adaptive membership function is used to calculate the dynamic benchmark adaptive membership value of each parameter.

[0086] Risk situation entropy is obtained by multi-source feature tensor fusion based on each parameter and its corresponding dynamic benchmark adaptive membership value.

[0087] In existing technologies, membership values ​​are directly calculated using trigonometric functions, trapezoidal functions, or Gaussian functions, and their parameters are set manually or empirically. Once the parameters are set, they cannot be automatically adjusted according to real-time data fluctuations, which can easily lead to overfitting or failure. Furthermore, all parameters use the same membership function without considering the distribution characteristics of different variables. Therefore, although existing methods for calculating membership values ​​are simple and easy to implement, they lack flexibility and intelligence.

[0088] The specific methods for obtaining the dynamic baseline adaptive membership value of each parameter in the calculation of the state parameters of mechanical and electrical equipment using the dynamic baseline adaptive membership function include:

[0089] Use linear functions, parabolic functions, and S-curve functions as reference functions;

[0090] The benchmark function is dynamically selected based on the statistics of each parameter; the statistics include standard deviation, skewness, and kurtosis.

[0091] Select the corresponding benchmark function based on the statistic. If the standard deviation is within the standard range, choose the linear function and calculate the membership value; if the skewness is within the standard range, choose the parabolic function and calculate the membership value; if the kurtosis is within the standard range, choose the S-curve and calculate the membership value.

[0092] If the standard deviation is greater than the standard range, select a membership value of 1; if the skewness is greater than the standard range, select a membership value of 1; if the kurtosis is greater than the standard range, select a membership value of 1.

[0093] If the standard deviation is less than the standard range, select a membership value of 0; if the skewness is less than the standard range, select a membership value of 0; if the kurtosis is less than the standard range, select a membership value of 0.

[0094] The weights of the statistics for each parameter are optimized using gradient descent, and then weighted and fused to obtain the dynamic baseline adaptive membership value.

[0095] Based on the specific values ​​of statistical parameters of mechanical and electrical equipment, including standard deviation, skewness and kurtosis, the most suitable benchmark function is selected, including linear function, parabolic function and S-curve function, as well as 0 or 1. The weight of the statistical parameters is optimized by combining gradient descent method, and finally the dynamic membership value is obtained by fusion.

[0096] Standard deviation, skewness, and kurtosis all have standard ranges, which are determined through expert experience.

[0097] The membership value is calculated by dynamically selecting a reference function based on the distribution characteristics of the state parameters of mechanical and electrical equipment, thus avoiding the limitations of a single function.

[0098] Standard deviation measures the dispersion of data (the uniformity of data distribution). When the data distribution is uniform (normal standard deviation), the degree of deviation is linearly related to the membership degree. For example, if a temperature parameter fluctuates uniformly under stable operating conditions, a linear function can smoothly quantify its deviation from the baseline. If the standard deviation is too large (e.g., sensor noise), the data is unreliable, and a membership degree of 1 (extreme anomaly) is directly assigned. If the standard deviation is too small (e.g., sensor failure), the data has no fluctuation, and a value of 0 is assigned (not involved in decision-making).

[0099] When the distribution is symmetrical (with normal skewness), the degree of deviation has a quadratic relationship with the membership degree. For example, in the normal distribution of vibration frequency, the greater the deviation from the mean, the faster the membership degree decreases, which conforms to the symmetry characteristic of a parabolic function. If the skewness is too large (severely skewed distribution), it indicates that the data distribution is abnormal, and a value of 1 should be directly assigned (this needs to be handled first). If the skewness is too small (the distribution is too symmetrical, such as due to artificial data correction), a value of 0 should be assigned (the data may be distorted).

[0100] When the kurtosis in the state parameters of mechanical and electrical equipment meets the standard range, the S-curve function is used to calculate the dynamic baseline adaptive membership value because the S-curve can smoothly reflect the gradual changes in the equipment state and adapt to the fluctuations in normal operation of the equipment. When the kurtosis is greater than the standard range, it indicates that the equipment has significant abnormal fluctuations or extreme behavior. In this case, the membership value is set to 1, indicating that the equipment is in an abnormal state and needs to be dealt with first. Conversely, if the kurtosis is less than the standard range, it indicates that the equipment is operating smoothly without abnormal fluctuations. The membership value is set to 0, and the system can ignore this state to avoid unnecessary intervention.

[0101] By integrating multidimensional statistics such as standard deviation, skewness, and kurtosis, the most suitable membership function type (such as a straight line, parabola, or S-curve) is dynamically selected, and gradient descent is introduced to optimize the weights, thus achieving adaptive calculation of membership values ​​for different state parameters. This method not only improves the accuracy of identifying the state of complex and nonlinear equipment, but also has stronger robustness and real-time response capabilities.

[0102] In existing methods, a fixed communication resilience strategy is selected, and most systems are preset to single-channel or dual-channel transmission. The fixed communication resilience strategy cannot sense the device's operating status, resulting in excessive communication in low-risk conditions and insufficient communication strength in high-risk conditions. A one-size-fits-all configuration may waste bandwidth in low-risk conditions and lack necessary redundancy in high-risk conditions.

[0103] Specific methods for dynamically selecting communication resilience strategies using risk posture entropy values ​​in risk selection rules include:

[0104] Communication resilience strategy selection: The risk situation entropy value comprehensively reflects the degree of danger of the current operating condition; when the risk situation entropy value is within the adaptive confidence boundary, dual-channel redundancy verification is initiated; when the risk situation entropy value is greater than the adaptive confidence boundary, three-channel mirror transmission is activated (adding power line carrier channels); when the risk situation entropy value is less than the adaptive confidence boundary, single-channel economic mode is adopted.

[0105] Based on the magnitude of the risk situation entropy value, the system automatically selects different communication resilience strategies to ensure that the equipment is fully protected in high-risk situations and can operate efficiently in low-risk situations.

[0106] Dual-channel redundancy verification mode: When the risk situation entropy value is within a safe range (i.e. within the adaptive confidence boundary), the system starts the dual-channel redundancy verification mode; at this time, the device is still operating normally, and the dual-channel redundancy verification ensures the stability of data transmission and avoids malfunctions caused by sensor failures or data transmission problems.

[0107] Three-channel mirror transmission mode: If the risk situation entropy value exceeds the adaptive confidence boundary, it indicates that the mechanical and electrical equipment is in a high-risk state (e.g., excessively high temperature or abnormal voltage). The system will activate three-channel mirror transmission. This method adds an extra data transmission channel (e.g., using power line carrier channel for data transmission) to ensure that critical safety information can be transmitted quickly and reliably and emergency stop control can be triggered when the equipment is in an emergency.

[0108] Single-channel economic mode: When the mechanical and electrical equipment is in good operating condition and the risk situation entropy value is low (such as normal wind speed and stable voltage), the system will adopt the single-channel economic mode; at this time, there is less redundancy, the system efficiency is higher, and unnecessary resource consumption is reduced.

[0109] Single-channel economic mode, including Industrial Ethernet; dual-channel redundancy check, including Process Ethernet, LoRa or Power Line Carrier; triple-channel mirrored transmission, including Industrial Ethernet, LoRa and Power Line Carrier.

[0110] The communication resilience strategy automatically adjusts in real time according to the risk situation entropy value, without relying on fixed communication, and adapts to the changing field environment; when the risk is high, a three-channel guarantee (such as power line carrier) is introduced, and when the risk is low, it switches to a single channel to save bandwidth and energy consumption; the communication resilience strategy is no longer passive, but actively adjusts based on the "risk situation entropy value of the actual state parameters of mechanical and electrical equipment", integrating perception and control logic.

[0111] In existing heartbeat protocols, the sending frequency of heartbeat packets is usually preset or based on a fixed time interval. Some systems simply set a fixed time interval to send heartbeat packets (e.g., sending a heartbeat packet every 10 seconds). In some systems, the sending frequency is selected based on a threshold change in sensor data (such as temperature, pressure, etc.). Existing methods often ignore the "actual state" of the device and manage the heartbeat packet frequency through simple static rules or timing strategies.

[0112] Under the selected communication resilience strategy, the specific methods for adjusting the heartbeat packet transmission frequency using the adaptive heartbeat synchronization strategy include:

[0113] Retrieve risk situation entropy values ​​from F historical time periods and calculate their moving average; set a base frequency based on the moving average; this represents the heartbeat packet transmission frequency of mechanical and electrical equipment under no-risk, low-risk, or high-risk conditions; the formula for the base frequency is: Base Frequency = *Sliding average+ ,in, and To adjust the parameters and ensure that the reference frequency is within a reasonable range;

[0114] The baseline frequency is adjusted based on changes in historical risk situation entropy values. When the risk situation entropy value increases over a period of time, the baseline frequency will also increase accordingly. Specifically, the baseline frequency is proportional to the moving average over a period of time and has a decay factor to smooth the changes and ensure that the frequency is within a reasonable range.

[0115] The initial transmission frequency is constructed by mapping the current risk situation entropy value to the base frequency; if the current risk situation entropy value is less than the adaptive confidence boundary, the initial transmission frequency is 50% of the base frequency to reduce communication resource consumption.

[0116] If the current risk situation entropy value is within the adaptive confidence boundary, the initial transmission frequency is an exponential increase of the base frequency, as shown in the formula: ,in, This is the initial transmission frequency. Indicates the reference frequency. Indicates response sensitivity. Control the steepness of the curve, This represents the risk situation entropy value;

[0117] If the current risk situation entropy value is greater than the adaptive confidence boundary, the initial transmission frequency is 150% of the base frequency;

[0118] The initial transmission frequency is dynamically corrected by incorporating the rate of change of the current risk situation entropy value, thus obtaining the transmission frequency.

[0119] The channel robustness index is a comprehensive indicator used to dynamically evaluate the reliability of communication links; communication links refer to dual-channel redundancy check mode, three-channel mirror transmission mode, and single-channel economic mode.

[0120] The frequency of heartbeat packet transmission is adjusted based on changes in the risk situation entropy value. When mechanical and electrical equipment is in a low-risk state, the frequency of heartbeat packet transmission is reduced, while when the equipment is in a high-risk state, the frequency is increased to ensure timely feedback on the equipment status at critical moments. This is more intelligent than the "fixed frequency" transmission in existing methods and can effectively adapt to real-time changes in equipment operating conditions.

[0121] The specific methods for calculating the packet loss rate based on the heartbeat packet transmission frequency and constructing a channel robustness index based on the packet loss rate and transmission jitter include:

[0122] Packet loss rate represents the proportion of heartbeat packets that are not successfully received during communication out of the total number of packets sent.

[0123] The packet loss rate is calculated by counting the total number of heartbeat packets sent within a time period and the number of unacknowledged responses received; the total number of heartbeat packets is the product of the time period and the heartbeat packet sending frequency.

[0124] The method for obtaining the number of heartbeat packets that did not receive an acknowledgment response is as follows: the sending frequency and sending time of the heartbeat packets are known; after each heartbeat packet is sent, it waits for an acknowledgment response (including acknowledgment number and sending timestamp) from the other end (such as the server). If no acknowledgment response is received within the set timeout period (such as 3 times RTT, i.e., round-trip time), it is determined that the heartbeat packet was not successfully received. The total number of heartbeat packets that did not receive an acknowledgment response during the accumulated time period is the number of heartbeat packets that did not receive an acknowledgment response.

[0125] Transmission jitter is defined as the standard deviation of the arrival time of heartbeat packets within a time period.

[0126] The channel robustness index is calculated based on packet loss rate and transmission jitter, using the following formula: , For channel robustness index, For packet loss rate, For heartbeat packet transmission jitter, For maximum transmission jitter, and These are the weights for packet loss rate and packet skipping jitter, respectively.

[0127] Risk decisions are made based on risk situation entropy, channel robustness index, and mechanical and electrical equipment status parameters, and the emergency shutdown confidence level is calculated. Specific methods for constructing the comprehensive risk trigger probability based on the emergency shutdown confidence level and channel robustness index include:

[0128] Step aa1, Risk Decision: Input Nodes: The risk situation entropy value, channel robustness index, and mechanical and electrical equipment status parameters are used as input nodes;

[0129] Output node: The output node represents the emergency stop confidence level;

[0130] Step aa2: Set the comprehensive risk trigger probability of mechanical and electrical equipment based on the emergency stop confidence level and the channel robustness index;

[0131] The failure risk of the channel robustness index is calculated by comparing it with the theoretical maximum index of channel robustness; the failure rate of the channel robustness index is then calculated using the failure risk of the channel robustness index.

[0132] The overall risk trigger probability is calculated by weighting the emergency shutdown confidence level and the channel failure rate.

[0133] Some existing systems make judgments on risk assessments based on simple triggering conditions (such as fixed risk situation entropy thresholds); they lack comprehensive and sufficient consideration of complex operating conditions.

[0134] Specific methods for setting multi-factor decision-making based on risk trigger probability include:

[0135] Level 1 Emergency Stop: When the overall risk trigger probability is higher than 0.7, it indicates that the mechanical and electrical equipment is in an extremely high-risk state and an emergency stop operation is performed. At this time, an emergency stop command is sent to cut off the power to the entire system and lock the mechanical devices to ensure equipment safety.

[0136] Level 2 slow stop: When the overall risk trigger probability is less than 0.7 and greater than 0.5, it indicates that the mechanical and electrical equipment is in a high-risk state, but it is not necessary to immediately execute an emergency stop; at this time, a phased speed reduction (such as reducing the equipment running speed) and partial power cut-off are initiated to reduce the equipment load and ensure a safe transition to the shutdown state;

[0137] Manual review mode: When the emergency stop confidence level is lower than 0.5, and the risk situation entropy value of the mechanical and electrical equipment exceeds the adaptive confidence boundary E times consecutively in the past H time steps, it indicates that the current state is abnormal and manual intervention is required; at this time, the system enters the manual review mode, requiring operators to manually review and judge.

[0138] By comprehensively considering multiple factors (such as emergency shutdown confidence and channel robustness index) to calculate the overall risk trigger probability, this multi-dimensional assessment method is more comprehensive than the existing single-factor judgment and can more accurately reflect the overall risk status of the equipment. This can more accurately assess the true risk status of the equipment, thereby avoiding premature or late triggering of safety measures and enhancing the reliability of the system.

[0139] Specific methods for compensating for risk situation entropy, channel robustness index, and emergency shutdown confidence include:

[0140] Damping force is a counterforce that slows down or inhibits the motion of an object. It usually occurs during the vibration or motion of an object or system. Damping force is essentially a manifestation of energy dissipation. It converts the kinetic energy of an object into heat energy, sound energy, or other forms of energy, thereby reducing the vibration amplitude of the system and making the system tend to be stable.

[0141] Damping force absorbs vibration energy, reduces the vibration amplitude and energy dissipation rate of mechanical systems, thereby suppressing abnormal vibration, reducing mechanical shock and structural fatigue, and directly reducing the risk of equipment failure caused by vibration (such as mechanical component breakage, bearing wear, poor electrical contact, etc.).

[0142] Active dampers, passive dampers, and failure dampers are installed in mechanical and electrical equipment. The process damping coefficient is calculated by collecting the active damping coefficient, passive damping coefficient, and failure damping coefficient from the active dampers, passive dampers, and failure dampers. The process damping coefficient = a * active damping coefficient + b * passive damping coefficient + c * failure damping coefficient, where a, b, and c represent a active dampers, b passive dampers, and c failure dampers installed in the mechanical and electrical equipment, respectively.

[0143] For the risk situation entropy value, the process damping coefficient is used for compensation to obtain the compensated risk situation entropy value. The formula is: Compensated risk situation entropy value = Process damping coefficient * Risk situation entropy value;

[0144] Zigbee modules are deployed in mechanical and electrical equipment. The RSSI values ​​at L locations are recorded using a mobile receiver. The average RSSI value is calculated based on the RSSI values ​​at L locations. The average RSSI value is then used to compensate for the channel robustness index, resulting in a compensated channel robustness index.

[0145] To compensate for the emergency shutdown confidence level, the risk situation entropy value and the channel robustness index are used to obtain the compensated emergency shutdown confidence level.

[0146] The specific methods for calculating the compensation effect index based on the compensation risk situation entropy value, compensation channel robustness index, compensation emergency stop confidence level, and mechanical and electrical equipment operating state feature vector include:

[0147] The weights for calculating the compensating risk situation entropy value are obtained using the exponential decay method, as shown in the formula: ,in, To compensate for the weight of the risk situation entropy value, To compensate for the risk situation entropy value, This is the upper limit of the adaptive confidence boundary. To adjust the parameters, This represents the risk situation entropy value.

[0148] The weights of the compensation channel robustness index are calculated using the Sigmoid function method, and the formula is as follows: ,in, To compensate for the weight of the channel robustness index, For adjustment coefficients, To compensate for the channel robustness index, 60% is the benchmark value (i.e., the inflection point) for compensating for the channel robustness index.

[0149] The weights for compensating for emergency stop confidence are calculated using the rate of change method, and the formula is as follows: ,in, To compensate for the weighting of the emergency stop confidence level, For emergency shutdown confidence level, For adjustment coefficients, The confidence level of the emergency shutdown at the previous time point;

[0150] Among them, the compensation risk situation entropy value usually reflects the safety risk status of the equipment; in reality, the safety risk of mechanical and electrical equipment does not change linearly; when the risk situation entropy value of mechanical and electrical equipment is high, its impact on the entire system will be more significant than when the compensation risk situation entropy value is low; the exponential decay method can capture this nonlinear characteristic well: as the risk situation entropy value increases, the rate of change of the weight gradually slows down;

[0151] The Sigmoid function has an S-shaped curve characteristic, which can achieve a smooth transition around a certain value (such as 60%) of the compensation channel robustness index. When it is below 60%, the weight gradually decreases; when it is above 60%, the weight gradually increases. This characteristic is very suitable for dealing with the weighting problem of the compensation channel robustness index. When the compensation channel robustness index is below 60%, its impact on the system gradually weakens, and when it is above 60%, the weight increases, reflecting the positive impact of the improvement of the compensation channel robustness index on the overall decision.

[0152] The rate of change method dynamically adjusts the weights by calculating the difference between the current confidence level of the compensated emergency stop and the confidence level of the compensated emergency stop at the previous time point. This method can effectively capture sharp changes in the confidence level of the compensated emergency stop and increase its weight when the confidence level changes significantly. Specifically, when the confidence level of the emergency stop fluctuates greatly, it indicates that the system state is changing drastically and may require a faster decision response, so increasing the weight is reasonable.

[0153] The comprehensive contribution value is obtained by multiplying the compensation risk situation entropy value, the compensation channel robustness index, and the compensation emergency shutdown confidence value with their corresponding weights. This part reflects the contribution of each key parameter to the system decision. The larger the weight, the more significant the parameter's impact on the system decision.

[0154] The total weight value is obtained by adding the weights of the compensation risk situation entropy value, the compensation channel robustness index, and the compensation emergency shutdown confidence value.

[0155] The overall evaluation value is obtained by comparing the comprehensive contribution value and the total weight value.

[0156] The deviation ratio is calculated by comparing the ratio of the mechanical and electrical equipment condition parameters to the mechanical and electrical equipment condition difference parameters; where the mechanical and electrical equipment condition difference parameters are the differences between the mechanical and electrical equipment condition parameters and the mechanical and electrical equipment condition standard parameters.

[0157] The compensation effect index is calculated using the total evaluation value and the deviation ratio.

[0158] Existing methods typically employ a one-time judgment mechanism, which executes certain operations, such as emergency stop, once the state becomes abnormal. However, it lacks the tracking and judgment of the long-term state of the equipment and may fail to identify ongoing risks after a short-term anomaly. If the equipment deteriorates and continues to experience problems, existing methods may fail to identify and execute emergency stop or other safety measures in a timely manner through continuous triggering mechanisms.

[0159] Specific ways to trigger risk decisions or multi-factor decisions based on the compensation effect index include:

[0160] If the compensation effect index is greater than If the compensation effect is unsatisfactory, a risk decision is triggered, the risk trigger probability is recalculated, and new factor trigger conditions are established based on the new risk trigger probability; among which, For channel robustness index, This is the adjustment coefficient;

[0161] If the compensation effect index is greater than J times consecutively This triggers a first-level emergency stop in multi-factor decision-making;

[0162] If the compensation effect index is less than or equal to If the compensation effect is normal, then monitoring will continue.

[0163] The triggering mechanism based on dynamic compensation effect and channel robustness index is more flexible than existing methods. It can adjust decisions according to the actual compensation effect, equipment status and changes in channel robustness, so as to make timely responses when the compensation effect is not good and reduce potential equipment damage and security risks.

[0164] In this embodiment, a feature vector of the operating state of mechanical and electrical equipment is established through the state parameters of the mechanical and electrical equipment. Sliding time window analysis and multi-dimensional feature extraction (statistics / frequency domain features) ensure the timeliness of data and solve the delay problem caused by traditional batch processing.

[0165] The adaptive confidence boundary is calculated using an LSTM model, and the dynamic baseline adaptive membership value of each mechanical and electrical equipment state parameter is calculated using a dynamic baseline membership function. The baseline function is intelligently selected based on the parameter statistical characteristics (standard deviation / skewness / kurtosis) to achieve adaptive adjustment of anomaly sensitivity.

[0166] The risk situation entropy value is not only based on the current state assessment, but also integrates a weighted model of dynamic membership function and multi-dimensional features, which effectively improves the accuracy and stability of the judgment.

[0167] By using an adaptive heartbeat synchronization strategy, the heartbeat frequency is dynamically adjusted, reducing the waste of communication resources while dynamically responding to the degree of danger, thus improving the system response efficiency. The channel robustness index not only considers the packet loss rate but also introduces transmission jitter, providing a more comprehensive assessment of data link stability.

[0168] Different levels of emergency stop strategies (Level 1 emergency stop, Level 2 gradual stop, and manual review) achieve "flexible shutdown" and avoid unnecessary economic losses caused by a one-size-fits-all emergency stop.

[0169] A comprehensive compensation method based on process damping compensation, communication RSSI compensation, and emergency shutdown confidence compensation can effectively repair system fluctuations caused by noise disturbances, false alarms, and missed alarms.

[0170] Ultimately, by using the compensation effect index, a closed-loop risk decision is triggered, enabling the system to have self-diagnosis, self-repair, and self-decision-making capabilities.

[0171] Example 2

[0172] Please see Figure 3 As shown, parts not described in detail in this embodiment are described in Embodiment 1. A real-time monitoring-based emergency stop control system for mechanical and electrical equipment is provided, including:

[0173] The risk identification subsystem: Drives a heterogeneous sensor array to synchronously acquire state parameters of mechanical and electrical equipment in the time domain; analyzes the state parameters of the mechanical and electrical equipment using sliding time window analysis to establish a feature vector of the operating state of the mechanical and electrical equipment; backtracks to extract the feature vectors of the operating state of the mechanical and electrical equipment from N historical time steps and inputs them into an LSTM model to generate adaptive confidence boundaries; uses a dynamic benchmark adaptive membership function to calculate the dynamic benchmark adaptive membership value of each parameter in the state parameters of the mechanical and electrical equipment; and performs multi-source feature tensor fusion based on each parameter and its corresponding dynamic benchmark adaptive membership value to obtain the risk situation entropy value.

[0174] Communication Assurance Subsystem: The communication resilience strategy is dynamically selected using the risk situation entropy value in the risk selection rules. Under the selected communication resilience strategy, the heartbeat synchronization strategy is adopted to adjust the heartbeat packet transmission frequency. The packet loss rate is calculated based on the heartbeat packet transmission frequency. The channel robustness index is constructed based on the packet loss rate and transmission jitter.

[0175] Decision execution subsystem: Makes risk decisions based on risk situation entropy, channel robustness index, and mechanical and electrical equipment status parameters, and calculates the emergency stop confidence level; constructs a comprehensive risk trigger probability based on the emergency stop confidence level and channel robustness index; and sets multi-factor decisions based on the risk trigger probability.

[0176] The optimization compensation subsystem compensates for the risk situation entropy, channel robustness index, and emergency shutdown confidence; it calculates the compensation effect index based on the compensated risk situation entropy, compensated channel robustness index, compensated emergency shutdown confidence, and mechanical and electrical equipment operating status feature vector; and it triggers risk decisions or multi-factor decisions based on the compensation effect index.

[0177] Example 3

[0178] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the operation mode of the above-mentioned emergency stop control method for mechanical and electrical equipment based on real-time monitoring.

[0179] Since the electronic device described in this embodiment is the electronic device used to implement the emergency stop control method for mechanical and electrical equipment based on real-time monitoring in the embodiments of this application, those skilled in the art can understand the specific implementation method and various variations of the electronic device in this embodiment based on the emergency stop control method for mechanical and electrical equipment based on real-time monitoring described in the embodiments of this application. Therefore, how the electronic device implements the method in the embodiments of this application will not be described in detail here. Any electronic device used by those skilled in the art to implement the emergency stop control method for mechanical and electrical equipment based on real-time monitoring in the embodiments of this application falls within the scope of protection of this application.

[0180] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0181] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for users of ordinary technical skills, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for emergency stop control of mechanical and electrical equipment based on real-time monitoring, characterized in that, include: Step SS1: Drive the heterogeneous sensor array to synchronously acquire the status parameters of mechanical and electrical equipment in the time domain; Sliding time window analysis is used to analyze the state parameters of mechanical and electrical equipment and establish the operating state feature vector of mechanical and electrical equipment. The operating state feature vector of mechanical and electrical equipment at N historical time steps is extracted and input into the LSTM model to generate adaptive confidence boundaries. The dynamic benchmark adaptive membership function is used to calculate the dynamic benchmark adaptive membership value of each parameter in the state parameters of mechanical and electrical equipment. Based on each parameter and its corresponding dynamic benchmark adaptive membership value, multi-source feature tensor fusion is performed to obtain the risk situation entropy value. Step SS2: Dynamically select a communication resilience strategy using the risk situation entropy value in the risk selection rules. Under the selected communication resilience strategy, adopt an adaptive heartbeat synchronization strategy to adjust the heartbeat packet transmission frequency. Calculate the packet loss rate based on the heartbeat packet transmission frequency, and construct a channel robustness index based on the packet loss rate and transmission jitter. Step SS3: Make risk decisions based on the risk situation entropy value, channel robustness index, and mechanical and electrical equipment status parameters, and calculate the emergency stop confidence level; construct a comprehensive risk trigger probability based on the emergency stop confidence level and the channel robustness index. Set multi-factor decision-making based on the probability of risk triggering; Step SS5: Compensate for the risk situation entropy value, channel robustness index, and emergency shutdown confidence level; calculate the compensation effect index based on the compensated risk situation entropy value, compensated channel robustness index, compensated emergency shutdown confidence level, and mechanical and electrical equipment operating status feature vector; trigger risk decision-making or multi-factor decision-making based on the compensation effect index.

2. The emergency stop control method for mechanical and electrical equipment based on real-time monitoring according to claim 1, characterized in that, The specific methods for obtaining the risk situation entropy value include: The status parameters of mechanical and electrical equipment are acquired synchronously in the time domain through a heterogeneous sensor array. Sliding time window analysis is used to analyze the state parameters of mechanical and electrical equipment, and to establish the operating state feature vector of the mechanical and electrical equipment; wherein, the window length T and the sliding step size are set. ; Calculate statistical quantities and frequency domain characteristics of the mechanical and electrical equipment status parameters within each time window. The statistical quantities include mean, variance, skewness and kurtosis, and the frequency domain characteristics include low-frequency, mid-frequency and high-frequency quantities. Obtain the feature vectors of the operating state of mechanical and electrical equipment over N historical time steps, and input them into the LSTM model to generate adaptive confidence boundaries. The state parameters of mechanical and electrical equipment are normalized, and the dynamic benchmark adaptive membership function is used to calculate the dynamic benchmark adaptive membership value of each parameter. Risk situation entropy is obtained by multi-source feature tensor fusion based on each parameter and its corresponding dynamic benchmark adaptive membership value.

3. The emergency stop control method for mechanical and electrical equipment based on real-time monitoring according to claim 2, characterized in that, The specific method for obtaining the dynamic benchmark adaptive membership degree value of each parameter in the calculation of the state parameters of mechanical and electrical equipment using the dynamic benchmark adaptive membership degree function includes: Use linear functions, parabolic functions, and S-curve functions as reference functions; The benchmark function is dynamically selected based on the statistics of each parameter; the statistics include standard deviation, skewness, and kurtosis. Select the corresponding benchmark function based on the statistic. If the standard deviation is within the standard range, choose the linear function and calculate the membership value; if the skewness is within the standard range, choose the parabolic function and calculate the membership value; if the kurtosis is within the standard range, choose the S-curve and calculate the membership value. If the standard deviation is greater than the standard range, select a membership value of 1; if the skewness is greater than the standard range, select a membership value of 1; if the kurtosis is greater than the standard range, select a membership value of 1. If the standard deviation is less than the standard range, select a membership value of 0; if the skewness is less than the standard range, select a membership value of 0; if the kurtosis is less than the standard range, select a membership value of 0. The weights of the statistics for each parameter are optimized using gradient descent, and then weighted fusion is performed to obtain the dynamic baseline adaptive membership value.

4. The emergency stop control method for mechanical and electrical equipment based on real-time monitoring according to claim 3, characterized in that, The specific methods for dynamically selecting a communication resilience strategy using risk situation entropy values ​​in the risk selection rules, and adjusting the heartbeat packet transmission frequency using an adaptive heartbeat synchronization strategy under the selected communication resilience strategy, include: The risk selection rules include: when the risk situation entropy value is within the adaptive confidence boundary, dual-channel redundancy verification is activated; when the risk situation entropy value is greater than the adaptive confidence boundary, three-channel mirror transmission is activated; when the risk situation entropy value is less than the adaptive confidence boundary, single-channel economic mode is adopted. Dual-channel redundancy verification, three-channel mirror transmission, and single-channel economic mode are communication resilience strategies; Retrieve the risk situation entropy values ​​over F historical time periods and calculate their moving average; set the base frequency based on the moving average. An initial transmission frequency is constructed by mapping the current risk situation entropy value to a base frequency; if the current risk situation entropy value is less than the adaptive confidence boundary, the initial transmission frequency is 50% of the base frequency. If the current risk situation entropy value is within the adaptive confidence boundary, the initial transmission frequency is the exponential growth value of the base frequency; If the current risk situation entropy value is greater than the adaptive confidence boundary, the initial transmission frequency is 150% of the base frequency; The rate of change of the current risk situation entropy value is introduced to dynamically correct the initial transmission frequency, thus obtaining the transmission frequency.

5. The emergency stop control method for mechanical and electrical equipment based on real-time monitoring according to claim 4, characterized in that, The specific methods for calculating the packet loss rate based on the heartbeat packet transmission frequency and constructing the channel robustness index based on the packet loss rate and transmission jitter include: The packet loss rate is calculated by counting the total number of heartbeat packets sent within a time period and the number of unacknowledged responses received; the total number of heartbeat packets is the product of the time period and the heartbeat packet sending frequency. The method for obtaining the number of heartbeat packets that did not receive an acknowledgment response is as follows: the sending frequency and sending time of the heartbeat packets are known; after each heartbeat packet is sent, it waits for an acknowledgment response from the other end. If no acknowledgment response is received within the set timeout period, it is determined that the heartbeat packet was not successfully received. The total number of heartbeat packets that did not receive an acknowledgment response during the accumulated time period is the number of heartbeat packets that did not receive an acknowledgment response. Transmission jitter is defined as the standard deviation of the arrival time of heartbeat packets within a time period. The channel robustness index is calculated based on packet loss rate and transmission jitter.

6. The emergency stop control method for mechanical and electrical equipment based on real-time monitoring according to claim 5, characterized in that, The risk decision-making process is based on the risk situation entropy value, channel robustness index, and mechanical and electrical equipment status parameters, and the emergency shutdown confidence level is calculated. Specific methods for constructing the comprehensive risk trigger probability based on the emergency shutdown confidence level and the channel robustness index include: Step aa1, Risk Decision: Input Nodes: The risk situation entropy value, channel robustness index, and mechanical and electrical equipment status parameters are used as input nodes; Output node: The output node represents the emergency stop confidence level; Step aa2: Set the comprehensive risk trigger probability of mechanical and electrical equipment based on the emergency stop confidence level and the channel robustness index; The failure risk of the channel robustness index is calculated by comparing it with the theoretical maximum index of channel robustness; the failure rate of the channel robustness index is then calculated using the failure risk of the channel robustness index. The overall risk trigger probability is calculated by weighting the emergency shutdown confidence level and the channel failure rate.

7. The emergency stop control method for mechanical and electrical equipment based on real-time monitoring according to claim 6, characterized in that, The specific methods for setting multi-factor decision-making based on risk trigger probability include: Level 1 Emergency Stop: When the overall risk trigger probability is higher than 0.7, it indicates that the mechanical and electrical equipment is in an extremely high risk state and an emergency stop operation is performed; at this time, an emergency stop command is sent to cut off the power to the entire system and lock the mechanical devices; Level 2 slow stop: When the overall risk trigger probability is below 0.7 and above 0.5, it indicates that the mechanical and electrical equipment is in a high-risk state, but an immediate emergency stop is not required; at this time, a phased speed reduction and partial power cut-off are initiated. Manual review mode: When the emergency shutdown confidence level is lower than 0.5 and the risk situation entropy value exceeds the adaptive confidence boundary E times consecutively within H historical time steps, it indicates that the current state is abnormal and an early warning is issued.

8. The emergency stop control method for mechanical and electrical equipment based on real-time monitoring according to claim 7, characterized in that, The specific methods for compensating for the risk situation entropy value, channel robustness index, and emergency shutdown confidence include: Active dampers, passive dampers, and failure dampers are installed in mechanical and electrical equipment. The active damping coefficient, passive damping coefficient, and failure damping coefficient are collected through the active dampers, passive dampers, and failure dampers to calculate the process damping coefficient. The process damping coefficient is used to compensate for the risk situation entropy value, resulting in a compensated risk situation entropy value. Zigbee modules are deployed in mechanical and electrical equipment. The RSSI values ​​at L locations are recorded using a mobile receiver. The average RSSI value is calculated based on the RSSI values ​​at L locations. The average RSSI value is then used to compensate for the channel robustness index, resulting in a compensated channel robustness index. To compensate for the emergency shutdown confidence level, the risk situation entropy value and the channel robustness index are used to obtain the compensated emergency shutdown confidence level.

9. The emergency stop control method for mechanical and electrical equipment based on real-time monitoring according to claim 8, characterized in that, The specific methods for calculating the compensation effect index based on the compensation risk situation entropy value, compensation channel robustness index, compensation emergency stop confidence level, and mechanical and electrical equipment operating status feature vector include: The weights for compensating for risk situation entropy are calculated using the exponential decay method; the weights for compensating for channel robustness index are calculated using the sigmoid function method; and the weights for compensating for emergency shutdown confidence are calculated using the rate of change method. The comprehensive contribution value is obtained by multiplying the compensation risk situation entropy value, the compensation channel robustness index, and the compensation emergency shutdown confidence value and their corresponding weights. The total weight value is obtained by adding the weights of the compensation risk situation entropy value, the compensation channel robustness index, and the compensation emergency shutdown confidence value. The overall evaluation value is obtained by comparing the comprehensive contribution value and the total weight value. The deviation ratio is calculated by comparing the ratio of the mechanical and electrical equipment condition parameters to the mechanical and electrical equipment condition difference parameters; where the mechanical and electrical equipment condition difference parameters are the differences between the mechanical and electrical equipment condition parameters and the mechanical and electrical equipment condition standard parameters. The compensation effect index is calculated using the total evaluation value and the deviation ratio.

10. The emergency stop control method for mechanical and electrical equipment based on real-time monitoring according to claim 9, characterized in that, The specific methods for triggering risk decisions or multi-factor decisions based on the compensation effect index include: If the compensation effect index is greater than If the compensation effect is unsatisfactory, a risk decision is triggered, the risk trigger probability is recalculated, and new factor trigger conditions are established based on the new risk trigger probability; among which, For channel robustness index, This is the adjustment coefficient; If the compensation effect index is greater than J times consecutively This triggers a first-level emergency stop in multi-factor decision-making.

Images