Intelligent Optimization System for Safety Monitoring of High and Low Voltage Switchgear Based on End-to-Cloud Collaboration
By using a high- and low-voltage switchgear safety monitoring system that integrates on-site data and cloud context, evidence fragments and proof obligations are formed, solving the problem of evidence collaboration between the end side and the cloud, achieving the completeness and reliability of risk assessment, and reducing false alarms and miscontrols.
Patent Information
- Application Number
- CN202611260654.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-19
- Publication Date
- 2026-11-17
AI Technical Summary
In the current safety supervision of high and low voltage switchgear, it is difficult to effectively coordinate the evidence from the end side and the cloud, resulting in unclear risk assessment, false alarms and miscontrols, and a lack of rapid on-site and historical data verification mechanism.
The intelligent optimization system for safety supervision of high and low voltage switchgear, based on end-to-cloud collaboration, is adopted. It includes an end-side evidence processing module, a cloud-based evidence processing module, a proof obligation collaboration module, a danger time domain negotiation module, and a safety disposal decision module. By organizing on-site data to form evidence fragments, associating them with cloud-based context data, determining the necessary evidence domain, and generating proof obligations and danger time margins according to risk types, collaborative disposal decisions are made.
It improves the completeness and reliability of risk assessment, reduces false alarms and erroneous control, ensures that on-site control actions match the time domain of fault evolution, and enhances the reliability of switchgear safety supervision.
Smart Images

Figure CN123122013A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of safety supervision technology for high and low voltage switchgear, and more specifically, to an intelligent optimization system for safety supervision of high and low voltage switchgear based on end-to-cloud collaboration. Background Technology
[0002] High and low voltage switchgear, as control and protection equipment in power distribution systems, is subject to factors such as electrical condition, temperature rise, insulation condition, mechanical operation, internal environment, and on-site interlocking conditions for safe operation. With the application of on-site data acquisition devices and cloud management platforms, switchgear safety supervision is gradually forming a processing method that combines real-time end-side monitoring with cloud data analysis.
[0003] Current switchgear safety monitoring typically involves collecting on-site data such as phase current, phase voltage, temperature rise, humidity, partial discharge, mechanism operation, and cabinet door status from the end-side, and then analyzing and judging this data in the cloud based on historical curves, equipment ledgers, maintenance records, or operation and maintenance work orders. While this method provides a data foundation for anomaly identification, in actual handling, on-site and cloud-based evidence often operate independently, lacking a mechanism for collaborative verification around the same risk statement. The end-side can promptly obtain on-site physical quantities and local hardware status, but it struggles to quickly verify historical operation, cabinet topology, and operation and maintenance scenarios. The cloud can provide historical references and contextual information, but it cannot replace on-site verification of electrical legality such as circuit breaker location, grounding switch status, cabinet door status, live display status, and interlocking status. If the key evidence required for a risk type is not clearly listed, and unmet evidence domains are not transformed into feedback-based proof obligations, risk assessment easily remains at the level of alarm values or empirical rules, and the closure of pre-control evidence is unclear. On the other hand, different risks evolve at different speeds. If the division of labor for handling is not determined by combining the leeway of danger and the time required for end-to-end cloud collaboration, there may be situations where control is implemented before the evidence is closed, or where evidence is still waiting to be supplemented in the cloud when there is insufficient time for danger.
[0004] Therefore, it is necessary to propose an intelligent optimization system for the safety supervision of high and low voltage switchgear based on end-to-cloud collaboration, so as to establish a continuous collaborative processing flow between risk declaration formation, evidence supplementation, danger time domain judgment and on-site action adjudication. Summary of the Invention
[0005] To address the shortcomings of existing technologies, the purpose of this invention is to provide an intelligent optimization system for safety monitoring of high and low voltage switchgear based on end-to-cloud collaboration.
[0006] To achieve the above objectives, the present invention provides the following technical solution: The intelligent optimization system for safety supervision of high and low voltage switchgear based on edge-cloud collaboration includes an edge evidence processing module, a cloud evidence processing module, a proof obligation collaboration module, a danger time domain negotiation module, a safety disposal adjudication module, and a strategy correction module. The end-side evidence processing module is used to organize the switch cabinet field data into evidence fragments, and based on the evidence fragments, form a first risk statement carrying the end-side evidence domain that has been satisfied and the cloud evidence domain that has not been satisfied; The cloud-based evidence processing module is used to receive the first risk statement and evidence fragments, and after associating them with cloud context data, form a second risk statement carrying the evidence domain that has been satisfied in the cloud and the evidence domain that still needs to be verified on the end side. The proof obligation coordination module is used to determine the necessary evidence domain according to the risk type, generate proof obligations based on the evidence domains not satisfied in the first risk statement or the second risk statement, receive proof obligation feedback, and obtain proof support when the proof obligation involves multiple evidence fragments, so as to form a proof state by combining the closure of the proof obligation corresponding to the necessary evidence domain. The danger time domain negotiation module is used to convert risk characteristics into risk coordinates, calculate the danger time margin based on the risk coordinates, and generate the end-to-cloud division of labor results by combining the end-to-cloud collaborative time consumption. The safety handling decision module is used to determine the permissible action level based on the proof status, danger time margin, and local hard status, and to control the execution of on-site control actions according to the permissible action level and the local hard status verification results. The strategy correction module is used to correct the proof obligation parameters, danger time domain parameters, and action application conditions based on the event closure record after the risk event ends.
[0007] Furthermore, the end-side evidence processing module is used to organize continuous sampled values in the field data into statistical segments and event-type data into event evidence segments; the statistical segments include the start and end times of the window, the mean within the window, the trend of change, and the peak value, and the event evidence segments include the event trigger time and continuous data before, during, and after the trigger.
[0008] Furthermore, the end-side evidence processing module is also used to determine the credibility of the evidence fragment. The credibility of the evidence is determined based on the interval from the time of completion of data collection to the time of participation in the judgment, the sensor state coefficient, and the local hard state consistency coefficient. The local hard state consistency coefficient is obtained by comparing the circuit breaker position, grounding switch state, cabinet door state, live display state, and lockout state at the corresponding time of the evidence fragment.
[0009] Furthermore, the first risk statement includes the risk object, risk type, risk location, triggering evidence fragment, evidence domain that has been satisfied, evidence domain that has not been satisfied, proof items to be verified in the cloud, and conditions for the statement to be valid; among them, the content obtained on-site and meeting the credibility requirements is listed as evidence domain that has been satisfied, and the content that still needs to be verified in conjunction with historical operating curves, cabinet topology, operation and maintenance scenarios or comparison with similar equipment is listed as evidence domain that has not been satisfied.
[0010] Furthermore, the cloud-based evidence processing module is used to form a second risk statement based on the first risk statement and the evidence fragment; the cloud-based context data includes at least one of historical operating curves, primary wiring topology of cabinets, equipment ledgers, maintenance work orders, repair records, and control action feedback records; the second risk statement includes evidence domains that have been satisfied in the cloud and evidence domains that still require end-side verification.
[0011] Furthermore, the proof obligation coordination module is used to determine the necessary evidence domain according to the risk type. The necessary evidence domain is written in the form of a configuration file. The records in the configuration file include the risk type, necessary evidence domain, evidence domain attribution side, observable, triggering condition, judgment relationship, and whether the evidence domain belongs to the pre-control necessary evidence. When the pre-control necessary evidence is not closed, the security disposal decision module will not upgrade the corresponding risk statement to an executable risk conclusion.
[0012] Furthermore, the proof obligation coordination module is also used to generate proof obligations for evidence domains that have not yet been satisfied, and to determine the proof obligation feedback based on the support, rebuttal, or undeterminable results of the feedback from the end side or cloud regarding the proof obligation; when the proof obligation involves multiple evidence fragments, the proof support degree is obtained by combining the support, rebuttal, or undeterminable results of each evidence fragment for the object of proof and the credibility of the evidence, and a proof status is formed for the security handling and adjudication module to call based on the proof obligation feedback, proof support degree, and whether the proof obligation corresponding to the necessary evidence domain is closed.
[0013] Furthermore, the danger time domain negotiation module is used to convert the risk characteristics corresponding to the risk statement and increasing with the degree of danger into risk coordinates, determine the danger time margin according to the rate of change of the risk coordinates, and calculate the end-to-cloud collaboration time based on the time reserved for communication round trip, cloud computing, proof feedback and control execution. When the danger time margin is insufficient to cover the end-to-cloud collaboration time, the end-to-cloud division of labor result for entering the end-side immediate handling process is generated. When the danger time margin allows the end-to-cloud round trip, the end-to-cloud division of labor result for entering the proof obligation supplementary proof process is generated.
[0014] Furthermore, the safety handling adjudication module is used to determine the permissible action level based on the proof status, dangerous time margin, and local hard status; wherein, the local hard status is used to determine the local electrical legality flag of the field control action, and the field control action is executed after being verified by the end-side hard status; when the control suggestion issued by the cloud is higher than the permissible action level, the end side only executes actions not exceeding the permissible action level and provides feedback on the reason for the downgrade; when the control suggestion conflicts with the local hard status, the end side refuses to execute.
[0015] Furthermore, the strategy correction module is used to form an event closed-loop record after the risk event ends. The event closed-loop record includes risk statement, proof obligation, proof feedback, danger time margin, end-to-cloud collaboration time, action level, end-side execution result, retest result, and manual confirmation result. The strategy correction module is also used to correct the proof obligation parameters, danger time domain parameters, and action applicability conditions based on the retest result and manual confirmation result.
[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. The edge-side evidence processing module organizes continuous sampled values and event-based data into evidence fragments, and forms a first risk statement based on the credibility of the evidence. The cloud-based evidence processing module then links historical operating curves, primary wiring topology of the cabinet, equipment ledgers, maintenance work orders, repair records, and control action feedback records to form a second risk statement. The proof obligation coordination module further determines the necessary evidence domains according to the risk type and transforms unmet evidence domains into proof obligations. It then combines proof feedback, proof support, and the closure of necessary evidence domains to form a proof status. This enables a correspondence between risk statements, evidence sources, matters to be proven, and disposal conditions, helping to improve the completeness of risk judgment criteria and reduce false alarms and miscontrols under conditions of missing key evidence. 2. The hazard time domain negotiation module converts risk characteristics into risk coordinates, determines the hazard time margin based on the rate of change of the risk coordinates, and calculates the end-cloud collaboration time consumption by combining the time required for communication round trips, cloud computing, proof feedback, and control execution. This generates the end-cloud task allocation result for entering the end-side immediate handling process or the proof obligation supplementation process. The safety handling adjudication module then determines the permissible action level based on the proof status, hazard time margin, and local hard status, and controls the execution of on-site control actions according to the permissible action level and the end-side hard status verification results. When the cloud control suggestion is higher than the permissible action level or conflicts with the local hard status, the end-side execution is downgraded or rejected. This ensures that supplementation, restricted handling, and control execution are matched with the fault evolution time domain and the on-site electrical legality, which helps improve the reliability of switchgear safety supervision and handling. Attached Figure Description
[0017] Figure 1This is a block diagram of a smart optimization system for safety monitoring of high and low voltage switchgear based on edge-cloud collaboration. Figure 2 This is a flowchart illustrating the implementation of the proof obligation collaboration module of this invention; Figure 3 This is a flowchart illustrating the implementation of the hazardous time-domain negotiation module of the present invention. Detailed Implementation
[0018] Example, refer to Figure 1 The intelligent optimization system for safety supervision of high and low voltage switchgear based on end-to-cloud collaboration in this embodiment includes an end-side evidence processing module, a cloud-based evidence processing module, a proof obligation collaboration module, a danger time domain negotiation module, a safety disposal adjudication module, and a strategy correction module. Mod1, End-side Evidence Processing Module: Collects on-site data from the switchgear and organizes the collected on-site values into evidence fragments that can be used for proof and judgment.
[0019] S11. On-site data acquisition: The end-side evidence processing module connects to sensors, status acquisition units, and secondary circuit signals inside and outside the switchgear. The collected data covers several categories, including electrical operation, thermal status, insulation status, mechanism action, cabinet environment, and local hard status.
[0020] The electrical operation data may include phase current, phase voltage, zero-sequence current, power factor, harmonic content, and load change rate; thermal status data may include busbar connection point temperature rise, circuit breaker contact temperature rise, drawer insertion point temperature rise, cable terminal temperature rise, and secondary terminal block temperature rise; insulation status data may include transient ground voltage signal, ultrasonic signal, high-frequency current signal, and partial discharge pulse phase; mechanical action data may include opening and closing coil current, auxiliary contact flipping time, opening and closing time, cabinet vibration response, and drawer advancement position; cabinet internal environment data may include cabinet internal temperature, cabinet internal humidity, fan operating status, dehumidifier operating status, and cabinet internal lighting changes; local hardware status data may include circuit breaker position, grounding switch status, cabinet door status, live display status, five-proof interlock status, and local / remote switching status. One or more of the above data can be selected according to cabinet type, compartment structure, sensor arrangement, and operation management requirements.
[0021] S12, Evidence Fragment Generation: After obtaining continuous sampled values at the endpoint, the data is first processed according to its rate of change. Quantities with slower changes, such as temperature rise, humidity, and average load, are not directly used as individual sampling points in subsequent verification. Instead, they are organized into statistical segments within a sliding time window. This segment retains the start and end times of the window and includes the mean, trend, and peak value within the window to reflect the state changes over a period of time.
[0022] For event-type data such as partial discharge pulses, cabinet door opening and closing, fan start / stop, dehumidifier start / stop, circuit breaker opening and closing actions, and drawer position changes, the event trigger time is used as the center of the segment. The end-side extracts continuous data before, during, and after the trigger from the corresponding collected data to form an event evidence segment. This segment is written with the event type, trigger time, relevant measurement points, original data index, and extracted features for subsequent use in proof obligations.
[0023] When generating evidence fragments, the local hard state at the same moment is also written in, serving as on-site constraints for subsequent proof and adjudication. For opening and closing action events, the event evidence fragment records the moment the control command is issued, the moment the coil current starts, the moment the auxiliary contact flips, the moment the main circuit current is established, the circuit breaker position change, and the interlocking status. For cabinet door opening and closing events, the event evidence fragment records the cabinet door opening and closing status, the live display status, the grounding switch status, the five-prevention interlocking status, and the corresponding maintenance tag input status.
[0024] S13. Calculate the credibility of the evidence: The end-side evidence processing module calculates the first according to the following formula. Credibility of each piece of evidence: in, For the first The credibility of each piece of evidence; The interval between the completion of the collection of this evidence fragment and the time when it is used for judgment; This is the effective time constant for this type of data. During commissioning or historical stable operation segments, the duration for which this type of data recovers from a single change to a stable band and remains stable is read, and the more conservative duration is written into the configuration file. If historical segments are missing, they are filled in according to the equipment's factory technical documents, operating procedures, or status verification cycle. Shorter pulses are used for cabinet door status, live display status, and partial discharge pulses. Contact temperature rise, cabinet humidity and average load are measured using a longer period of time. .
[0025] This represents the sensor state coefficient. (Following the...) Each piece of evidence is written together. The terminal reads the power supply status, sampling clock record, verification message, communication retransmission record, and self-test return code of the corresponding sensor for that piece of evidence, and forms a sensor status flag accordingly. When all of the above items are normal, The value is set to 1. In cases of power supply fluctuations, clock drift, verification anomalies, or communication retransmission anomalies, the corresponding anomaly level in the configuration file will be deducted. If the result after deduction is lower than 0, it will be treated as 0; if it is higher than 1, it will be treated as 1. Each anomaly level is written during the commissioning configuration phase, specifically referring to the operating status definitions, self-test message return code meanings, and station communication quality requirements in the sensor manual.
[0026] The local hard state consistency coefficient is obtained by comparing the circuit breaker position, grounding switch status, cabinet door status, live display status, and interlocking status at the corresponding moment in the evidence segment. If there are no physical conflicts in the aforementioned hard states, Maintain a high value; if situations occur such as the live indicator being valid but the maintenance status not being established, the cabinet door being open but the lockout status indicating remote operation is allowed, or the grounding switch status contradicting the circuit breaker position, etc. Reduce. The determination relationship of hard state conflict is written into the end-side configuration file according to the primary wiring method of the switchgear, the five-prevention interlocking logic, and the field operation procedures.
[0027] S14. Statement of Risks at the End of the Process: The end-side evidence processing module generates a first risk statement based on the evidence fragments. The first risk statement includes the risk object, risk type, risk location, triggering evidence fragment, satisfied end-side evidence domain, unsatisfied cloud-based evidence domain, proof items pending cloud verification, and conditions for the statement's validity. Risk objects include at least one of the following: cabinet, compartment, drawer unit, circuit breaker unit, busbar connection point, cable terminal, and secondary circuit node. Risk types include at least one of the following: abnormal contact, insulation degradation, partial discharge, mechanical lag, accidental entry into a energized cabinet door, ventilation degradation, dehumidification failure, secondary state distortion, and sensor distortion.
[0028] In the risk statement on the edge side, content that has been obtained on-site and meets the corresponding credibility requirements is listed as the satisfied edge-side evidence domain; content that still needs to be verified in conjunction with historical operating curves, cabinet topology, maintenance scenarios, or comparison with similar equipment is listed as the unsatisfied cloud-based evidence domain. The conditions for the validity of the statement are given along with the risk statement, covering load range, cabinet door status, maintenance status, sensor status, and applicable time window. If the above conditions change, or the time window expires, the risk statement enters the state of pending re-verification.
[0029] Mod2, Cloud Evidence Processing Module: Receives risk statements and evidence fragments uploaded from the receiving end, supplementing historical, topological, and operational evidence that is difficult to obtain in a timely manner from the on-site end.
[0030] S21, Associate cloud context data: The cloud-based evidence processing module associates the evidence fragments uploaded from the terminal with the target switchgear's historical operating curves, primary wiring topology, equipment ledger, maintenance work orders, repair records, operation ticket status, power outage and restoration records, and control action feedback records, based on the cabinet number, compartment number, measurement point number, and event time. From the historical operating curves, periods with maintenance, sensor self-test anomalies, communication anomalies, control action disturbances, and manual blocking are removed, and the remaining periods are used as candidate health reference periods.
[0031] The historical operating curve elimination process is performed according to time stamps. If a certain time period contains markers for maintenance work orders, operation tickets, control actions, sensor self-test anomalies, or communication interruptions, that time period will not be included in the health reference candidate time period. If a certain time period contains records of forced fan operation, forced dehumidification operation, or load transfer, that time period will only be used to prove the effectiveness of the corresponding control actions and will not be used as a reference for natural health response. If the evidence fragment uploaded by the end side lacks cabinet number, measurement point number, or timestamp, the cloud evidence processing module will return an unassociatable feedback to the end side and request the end side to re-upload evidence fragments with complete indexes.
[0032] S22. Calculate the health response reference value: For risk types requiring historical response evidence, the cloud-based evidence processing module searches for segments in the health reference candidate time period that are close to the current operating condition. The current operating condition is used... express, This includes at least one of the following: load level, ambient temperature, internal humidity, fan status, dehumidification status, operating mode, and cabinet door status. (Cloud-based access to the first...) Calculate health response reference values for each target quantity: in, For the first The health response reference value of each target quantity under current operating conditions; The first candidate period for health reference Measured values of the target quantity for each sample; In order to adapt to the current working conditions A similar set of samples; For the first Weights for the similarity of working conditions among samples.
[0033] sample set The following steps are used to generate the data: First, the operating condition components, such as load, ambient temperature, cabinet humidity, operating mode, and control action status, are normalized according to the normal fluctuation range within the health reference candidate period. Then, the similarity between the candidate sample operating conditions and the current operating conditions is compared. Samples that meet the screening criteria are included. The closer the sample operating conditions are to the current operating conditions, the better. The larger the sample size, the better. If the target cabinet itself lacks sufficient health samples, the cloud will supplement the samples from switch cabinets with the same or similar equipment model, rated voltage, cabinet structure, installation environment, and sensor layout, and will then transfer the supplementary samples to the cloud. Set it to a value lower than the target cabinet's own sample value. If supplemented... If it is still empty, the cloud proof feedback for the target quantity is undetermined, and the health response reference value will not be calculated further.
[0034] In some embodiments, when the target quantity is the contact temperature rise... At least load level, ambient temperature, and fan status should be included; when the target quantity is the humidity decline curve, At least the following should be included: cabinet internal temperature, cabinet internal humidity, dehumidification status, and cabinet door status; when the target quantity is the mechanism's action time, At a minimum, this includes the operating power supply voltage, cabinet temperature, action type, and the status of the most recent maintenance. The above operating condition components can be reduced based on the actual sensor configuration, but components that cannot be used for evidence after reduction should be marked as not satisfying the evidence domain in the cloud-based risk statement.
[0035] S23. Form a cloud-based risk statement: The cloud-based evidence processing module generates a second risk statement based on health response reference values, cabinet topology, operation and maintenance context, and anomaly records of similar switchgear. The second risk statement includes the risk object, risk type, risk location, evidence domains already satisfied in the cloud, evidence domains still requiring end-side verification, matters to be proven at the end-side, and conditions for the statement's validity. When the cloud-based risk statement involves on-site electrical safety actions, it serves only as a risk statement awaiting end-side verification and is not directly used as the basis for on-site control execution.
[0036] The cloud-based evidence domains already satisfy at least one of the following: historical health response evidence domain, cabinet topology evidence domain, operation and maintenance scenario evidence domain, comparison evidence domain with similar switchgear, and control feedback evidence domain. Evidence domains still requiring end-side verification include at least one of the following: local hard state evidence domain, single physical response evidence domain, transient mechanism action evidence domain, field environment evidence domain, and sensor status evidence domain. The conditions for declaring validity include the operating condition range, operation and maintenance status range, topology applicability range, and time validity range of the historical references cited by the cloud.
[0037] Mod3, Proof Obligation Coordination Module: such as Figure 2 As shown, unclosed evidence is transformed into a verifiable question on the other side. When the edge makes a risk statement, the cloud is responsible for proving the history, topology, and operational context; when the cloud makes a risk statement, the edge is responsible for proving the on-site hard state, transient response, and sensor status.
[0038] S31. Determine the necessary evidence domain: After receiving the first or second risk statement, the proof obligation coordination module determines the necessary evidence domains according to the risk type. For abnormal contact, at least one physical response evidence, load correspondence evidence, historical health response evidence, and reliable sensor evidence are required. For partial discharge risk, at least local pulse evidence, phase evidence, adjacent cabinet synchronization evidence, and cabinet internal environment evidence are required. For mechanical hysteresis risk, at least action command, coil current, auxiliary contact flipping, and main circuit response evidence are required. For accidental entry into a energized cabinet door, at least cabinet door status, energized display status, grounding switch status, locked status, and maintenance scenario evidence are required. For ventilation degradation risk, at least fan operation status, temperature drop response, load status, and cabinet internal environment evidence are required. For dehumidification failure risk, at least dehumidification operation status, humidity drop response, cabinet door status, and ambient humidity evidence are required.
[0039] Necessary evidence domains are written in configuration files. Each record in the configuration file must include at least the risk type, necessary evidence domain, the domain's attribution, observables, triggering conditions, decision relationship, and whether the evidence domain is necessary evidence prior to control. When necessary evidence prior to control is not closed, the safety action decision module must not upgrade the corresponding risk statement to an executable risk conclusion; when non-necessary evidence prior to control is not closed, the risk statement can remain in observation or manual review status.
[0040] S32, Obligation to Prove: For evidence domains that have not yet been satisfied, the proof obligation collaboration module generates proof obligations. Proof obligations include proof number, risk statement number, target evidence domain, proof object, triggering condition, observation time window, target observation, judgment relationship, completion deadline, and feedback format. Proof objects are described as facts that can be directly verified by the end-user or cloud, such as whether abnormal temperature rise exceeds the health response reference under current operating conditions, whether a suspected partial discharge pulse appears synchronously in a neighboring cabinet, whether cabinet door opening is consistent with the maintenance scenario, and whether the auxiliary contact status is supported by the main circuit response.
[0041] The triggering conditions for the proof obligation are set according to the risk mechanism. For cases involving abnormal contact, the end-side trigger points are load increase, load decrease, or phase current step change, used to obtain response segments of temperature rise as the load changes. For cases involving partial discharge, the trigger point is the appearance of a suspected pulse in the target cabinet; when a synchronization pulse appears in an adjacent cabinet at a similar power frequency phase, it is also used as a trigger point to verify the pulse source. For cases involving mechanical lag, the trigger points are the opening command, closing command, or drawer push-in action. For cases involving accidental entry into a energized cabinet door, the trigger point is the change of the cabinet door status from closed to open, or a change in the energized display status.
[0042] The observation time window expands around the trigger point and includes data segments before, during, and after the event. The window length is written to the endpoint configuration file according to the risk type and sampling method; among them, transient events use shorter windows, while slowly changing events such as temperature rise and humidity use windows that can cover the change process.
[0043] S33. Execute proof and provide feedback: After receiving the proof obligation, the terminal side first checks whether the triggering conditions have been met. If the triggering conditions are met, the terminal side extracts the target observation from the circular buffer or the current acquisition window. If the triggering conditions are not met, the terminal side sets a one-time triggering condition to capture the observation time window when the next load step, fan start / stop, dehumidifier start / stop, cabinet door opening / closing, circuit breaker opening / closing action, or drawer pushing event occurs. After receiving the proof obligation, the cloud side calls the health response reference, cabinet topology, maintenance work order, operation ticket status, and control action feedback records for the corresponding time period for verification.
[0044] Edge-side feedback includes support, rebuttal, and indeterminate. If the edge can obtain the target observation and the target observation conforms to the determination relationship in the proof obligation, the feedback is supportive; if the target observation contradicts the object of proof, the feedback is rebuttal; if there is a lack of triggering events, insufficient credibility of evidence fragments, sensor self-test anomalies, or hard state conflicts, the feedback is indeterminate. Cloud-side feedback uses the same format. If the cloud cannot obtain a health response reference, the topology association is incomplete, the work order status is missing, or the operation ticket time does not match the event time, the feedback is indeterminate.
[0045] S34. Obtaining proof of support: When the burden of proof involves multiple pieces of evidence, the first piece of evidence shall be calculated according to the following formula. The degree of proof support for each proof obligation: in, For the first The degree of evidentiary support for each evidentiary obligation; The set of evidence fragments that participate in the determination of the burden of proof; For the first The weight of each piece of evidence in the burden of proof; For the first Each piece of evidence relates to the directional result of the object being proven; For the first The credibility of each piece of evidence.
[0046] The correspondence between evidence domains and risk types is pre-written into the configuration file. Within the same risk type, evidence that directly proves the fault mechanism has a higher weight, while indirect background evidence has a lower weight. Taking abnormal contact at a contact point as an example, the temperature rise response after load normalization, the temperature rise difference between phases, and the temperature rise drop curve are direct evidence, while the ambient temperature inside the cabinet and the temperature of adjacent compartments are background evidence. Taking partial discharge risk as an example, the pulse phase distribution, the amplitude of the target cabinet sensor, and the synchronization of adjacent cabinets are direct evidence, while the humidity inside the cabinet and the external switching records are background evidence.
[0047] The value is 1 when supporting the proof, -1 when refuting the proof, and zero when no decision can be made. If If the result is empty, or if all evidence fragments involved in the calculation are unusable due to low credibility, then the burden of proof is deemed undecidable.
[0048] When proving a judgment, if The feedback is supportive; if The feedback for the first case was a rebuttal; the feedback for the other cases was that it could not be determined. To support thresholds, The threshold for rebuttal, and satisfying The support threshold and rebuttal threshold are determined during the commissioning phase. Specifically, the following steps are taken: calculate the thresholds for the verified real risk segments, explained disturbance segments, and normal operation segments respectively. The system selects boundary values that ensure real-risk segments fall into the support side, explained perturbation segments fall into the rebuttal side, and normal operation segments do not trigger support judgments, and writes these values into the configuration file. If subsequent retesting results show that the boundaries are not compatible, the strategy correction module will make minor adjustments.
[0049] Mod4, Dangerous Time Domain Negotiation Module: such as Figure 3 As shown, the responsibilities of the edge and cloud are allocated according to the speed of fault evolution. When the margin of danger is short, the on-site edge side takes priority in freezing evidence and handling restrictions; when the margin of danger allows for round trips between the edge and cloud, the proof obligation coordination module takes priority in completing supplementary proof.
[0050] S41. Establish risk coordinates: The danger time-domain negotiation module converts different physical quantities into risk coordinates. The risk coordinate increases with the degree of danger, and the safety boundary is uniformly represented as... For parameters that exceed the upper limit, convert them using the following formula: in, For the first One risk coordinate; These are risk characteristic values that have been converted into increased hazards, such as temperature rise residuals, partial discharge pulse intensity increments, action time offsets, or humidity response residuals. This is the safety boundary value corresponding to the risk characteristic, and it is related to... The same measurement object is used. Write the allowable upper limit based on the equipment nameplate ratings, protection settings, insulation withstand requirements, manufacturer test data, or commissioning and acceptance records. For parameters exceeding the lower limit, first convert them into risk characteristic values that increase with the degree of danger, and then process them as described above. For example, convert insufficient condensation margin into an increased condensation risk. If... A value less than zero indicates that the current feature does not represent a dangerous increment, and the risk coordinate is used as zero in subsequent calculations.
[0051] Different types of risks Obtain it as follows. In case of contact abnormality, The positive deviation between the current contact temperature rise and the health response reference value can be taken, or the residual value where the temperature rise does not decrease after the load decreases. In the partial discharge risk, The positive deviation of the current pulse intensity statistical value relative to the background pulse statistical value under the same humidity and operating mode can be taken. In the risk of institutional hysteresis, The current operating time can be considered a positive deviation from the healthy operating time under the same operating power supply voltage and temperature conditions. In the risk of dehumidification failure, The humidity drop after the dehumidification process is insufficient. (The above...) Corresponding Use the same physical quantities or the same characteristic quantities to participate in risk coordinate transformation.
[0052] S42. Calculate the critical time margin: The danger time-domain negotiation module first performs window smoothing on the risk coordinates to obtain... Temperature rise risks are assessed using the mean or median value of a continuous sampling window; partial discharge risks are assessed using the pulse intensity statistics within an event window; and mechanism motion risks are assessed using the characteristic values of adjacent motion events. The subsequent calculation... The rate of change of each risk coordinate: in, For the first The rate of change of each risk coordinate; The smoothed value of the risk coordinates within the current calculation window; This represents the smoothed value of the risk coordinates within the previous calculation window; This is the time interval between two adjacent calculation windows or two adjacent event windows. For temperature rise risks, a continuous sampling window interval is used; for partial discharge risks, an adjacent event window interval is used; and for mechanism action risks, an adjacent action event interval is used.
[0053] The critical time margin for candidate faults is determined by the following formula: in, This represents the critical time margin for candidate faults. This is the set of risk coordinates corresponding to the candidate fault; This is the maximum observation time. If any risk coordinate has reached or exceeded the safety boundary, Recorded as 0. If there are continuously increasing risk coordinates, the time that first approaches the safety boundary is taken as the value. If all risk coordinates do not continue to increase, Recorded as . Write the endpoint cache coverage duration, inspection cycle, and risk type into the configuration file.
[0054] S43. Time consumed in computing edge-cloud collaboration: The dangerous time-domain negotiation module obtains the time required for one effective end-to-cloud collaboration according to the following formula: in, The time consumed by end-to-end cloud collaboration; An effective end-to-cloud collaboration consists of at least the following steps: communication round trip, cloud computing, proof feedback, and control execution reservation. For the first The time consumed in each step. The time consumed in the communication round trip is selected from the most recent communication records to cover most fluctuations; the time consumed in the cloud computing step is read from the historical processing records of the same risk type; the time consumed in the proof feedback step is determined according to whether the proof obligation needs to wait for natural operating events to determine; the time consumed in the control execution reserved step is determined according to the response records of the action mechanism or the factory technical documents.
[0055] The communication round-trip process can be recorded by the sending and confirmation times of the periodic heartbeat messages on the device side and the confirmation messages on the cloud. The cloud computing process can be obtained by the interval between receiving the risk statement on the cloud and generating the proof obligation or adjudication result. For the proof feedback process, if it pertains to an immediately verifiable matter, the time required for the device side to read the local cache or the cloud to read the context record is used; if it pertains to a matter awaiting a natural occurrence event, a conservative value of the interval between adjacent occurrences of the triggering event in the target cabinet's historical operation is used. The control execution reservation process is written according to the execution records of the action mechanism in the same model of equipment or the factory documentation.
[0056] S44. Generate the end-to-cloud task allocation results: like Candidate faults enter the edge-side immediate handling process, where the edge only performs handling actions that have been verified locally in terms of hard state and whose action level is restricted. If Candidate faults will then enter the supplementary proof obligation process. The deadline for completing the proof obligation is determined by the following formula: in, For the first The deadline for fulfilling the obligation of proof; This is the time limit coefficient, with a value greater than 0 and less than 1. Risks related to personal safety, rapid insulation, and delayed action are represented by the smaller coefficient. Slow temperature rise risk and ventilation decline risk adopt larger . The risk type is written during equipment commissioning, and the strategy correction module adjusts it based on the retest results. When... At that time, no waiting period is generated for proof, and the process directly enters the immediate processing procedure on the client side.
[0057] Mod5, Safety Handling Decision Module: Based on the proof status, hazard time margin, and local hard status, this module determines whether a risk statement should be upgraded, downgraded, converted, or placed under restricted control. On-site control actions must be executed after end-side hard status verification.
[0058] S51, Risk Statement Regarding Adjudication: If all necessary proof obligations for the same risk object and risk type are supported, the safety handling adjudication module upgrades the risk statement to an executable risk conclusion. An executable risk conclusion includes the risk object, risk type, evidence closure status, permissible action level, required continuous monitoring of the response quantity, and conditions for the conclusion's validity. If any necessary proof obligation is refuted, the risk statement is downgraded, converted, or revoked. If the proof obligation remains undetermined within the deadline, and Continue to seek further evidence or remain under observation; if Access will be restricted and immediate action will be taken.
[0059] When the proof obligations for abnormal contact are all supported, the response quantities that need to be continuously monitored in the enforceable risk conclusion include temperature rise residual, load change, and temperature rise fall curve. When the proof obligations for partial discharge risk are all supported, the response quantities that need to be continuously monitored include pulse phase distribution, pulse intensity, and synchronicity with adjacent cabinets. When the proof obligations for mechanical hysteresis risk are all supported, the response quantities that need to be continuously monitored include coil current characteristics, auxiliary contact flipping time, and main circuit response. When the proof obligations for the risk of accidental entry into a energized cabinet door are all supported, the response quantities that need to be continuously monitored include cabinet door status, energized display status, grounding switch status, and lockout status.
[0060] S52. Determine the permitted action level: The safety handling decision module determines the highest permissible action level using the following formula: in, The highest level of action currently allowed to be performed; The preset action level set includes evidence freezing, increasing sampling frequency, local alarm, low-risk environment adjustment, remote operation restriction, manual confirmation request, lockout prompt and forced lockout; For the first The time required for a level action to generate a valid response from the moment it is triggered; This serves as a mark of local electrical legality. For the first The degree to which a level of action disturbs subsequent diagnostic evidence; This represents the maximum perturbation allowed in the current state of evidence closure.
[0061] The response time is obtained from the historical execution records of this action in this cabinet or the same model actuator. If historical records are missing, the action response time in the factory technical document is used. The value is obtained by reading the circuit breaker position, grounding switch status, cabinet door status, live display status, five-proof interlock status, and local / remote switching status from the end side. If all the prerequisites for this action are met, the value is 1; otherwise, the value is 0.
[0062] Determined through the motion disturbance field. The motion disturbance field includes temperature disturbance, humidity disturbance, load disturbance, electromagnetic disturbance, and mechanism state disturbance. During configuration, first list the... The on-site quantity that a certain action would change is then compared with the evidence domain that is not yet closed in the current risk statement. If an action would change the key evidence domain on which the current risk statement relies, then... Increase; if an action does not change the key domain of evidence upon which the current risk statement relies, then Keep the value low. The value decreases as the evidence becomes closed; it decreases when the evidence is not yet closed and increases when the evidence is closed and the remaining time of danger is short. and The same motion disturbance level configuration is used to ensure that the two can be compared.
[0063] If the set of actions that meet the conditions is empty, then Take the lowest read-only action level This level includes evidence freezing, increasing sampling frequency, or local alarms, but does not execute control actions that change load, humidity, temperature, or machine status.
[0064] S53. Performing or refusing a control action: If the control recommendations issued from the cloud are higher than The endpoint only executes no more than The system will take action and provide feedback on the reason for the downgrade. If the control suggestion conflicts with the local hard state, the endpoint will refuse to execute and upload hard state evidence fragments, the reason for refusal, and the current remaining danger time. After the endpoint refuses to execute, the cloud evidence processing module will re-associate the operation and maintenance context and topology state, and the proof obligation collaboration module will regenerate the proof matters or transfer them to manual review.
[0065] After the action is completed on the end side, the action trigger time, action level, local hardware status before execution, response quantity after execution, and execution result are uploaded to the cloud. If the action involves fan start / stop, dehumidification control, or load adjustment, a disturbance flag is written in the corresponding time period. The disturbance flag indicates that the action may affect the temperature rise, humidity, or load evidence domain.
[0066] When subsequent proof obligations invoke data from that time period, disturbance markers must be checked first. Evidence fragments marked as affected by actions cannot be treated as natural operational responses, nor can they be used as evidence of uninterrupted natural responses to support risk proofs.
[0067] Mod6, Strategy Correction Module: Based on the retesting and confirmation results after the event ends, correct the proof obligation, danger time domain parameters, and action application conditions.
[0068] S61. Form a closed-loop record of events: After a risk event concludes, the strategy correction module records the risk statement number, risk object, risk type, triggering evidence fragment, proof obligation, proof feedback, hazard time margin, end-to-end cloud collaboration time, diagnostic lead side, action level, end-side execution result, retest result, and manual confirmation result. The retest result includes temperature rise changes, partial discharge changes, mechanism recovery status, cabinet environment changes, and local hardware status changes over a period after the event.
[0069] Closed-loop records are archived according to risk event numbers. Each proof obligation records the object of proof, the relevant evidence fragments, the degree of proof support, the feedback result, and whether it has been subsequently retested and confirmed. Each control action records the risk coordinates before the action, the risk coordinates after the action, the action disturbance field, and the action legality verification result. Manual confirmation results can come from on-site inspection records, maintenance records, test records, or fault handling forms.
[0070] S62. Modify the proof obligation parameter: If an event is confirmed as a real risk, the strategy correction module increases the weight of the evidence domain that plays a major role in the confirmation conclusion and can shorten the observation time window for similar risks. If an event is confirmed as a false alarm, the strategy correction module distinguishes the source of the false alarm based on the results of retesting and manual confirmation. When the source of the false alarm is an abnormal sensor status, the calculation rules for the corresponding sensor status coefficient are adjusted; when the source of the false alarm is a mismatch in historical health responses, the candidate health reference period is re-selected; when the source of the false alarm is a lag in the operation and maintenance situation, the requirements for the corresponding work order, operation ticket, or maintenance record in the proof obligation are increased; when the source of the false alarm is a disturbance in control actions, the degree of disturbance of the corresponding action is corrected. .
[0071] The adjustment of the proof obligation parameters is performed according to the event type. If a key evidence domain in a real risk event has provided supporting feedback multiple times before retesting and confirmation, then the level of that evidence domain in similar proof obligations is increased. If a particular piece of background evidence repeatedly leads to an excessively high level of support in a false alarm event, then the corresponding value of that piece of background evidence will be reduced. If a certain proof obligation returns to undeterminable status for an extended period, check whether its triggering conditions are unlikely to occur during the operation of the target cabinet. If the triggering conditions are unlikely to occur, change the proof obligation to an inspection and review item, or replace it with an observation that can be obtained by existing sensors under the same evidence domain.
[0072] S63. Correct the danger time domain and motion parameters: The strategy correction module will calculate the... Compare the actual time of risk deterioration or risk resolution observed during the retest. If the estimated time leeway for repeated occurrences of similar risks is too short, appropriately extend the corresponding timeframe for fulfilling the proof obligation; if the estimated time leeway for repeated occurrences of similar risks is too long, shorten the timeframe for fulfilling the proof obligation and increase the priority of immediate on-device handling. For control actions, if the target response does not change as expected after execution, adjust the response time of the action. degree of disturbance Or the conditions under which the action applies.
[0073] When correcting for hazardous time domain conditions, the first priority is to check whether the construction of the risk coordinates is consistent with the actual conditions observed during the retest. If the retest indicates that the abnormal temperature rise is mainly caused by load changes, then the temperature rise health response reference values should be re-selected to avoid insufficient load interpretation. The value is too high. If the retest shows that the partial discharge pulse originates from external common-mode interference, then the requirement for evidence of synchronicity between adjacent cabinets should be increased in the proof obligation for the corresponding risk type. If the retest shows that the lag in the mechanism's action deteriorates rapidly within a short period, then the smoothing window for risks related to mechanism action should be shortened. To reflect recent events more quickly. The above corrections are written to the configuration file after manual confirmation or confirmation by retest records.
[0074] Through the synergy of the six modules described above, the intelligent optimization system for safety supervision of high and low voltage switchgear based on end-to-cloud collaboration in this embodiment can incorporate real-time physical evidence from the end side, historical and operational evidence from the cloud, mutual proof obligations, and fault hazard time margin into the same handling chain. Risk statements are upgraded to executable risk conclusions after evidence closure; supplementary evidence is required when evidence is not yet closed but hazard time is sufficient; and limited handling is performed by the end side when hazard time is insufficient. This reduces false alarms and miscontrols under conditions of missing key evidence and avoids excessive waiting for cloud processing when faults evolve rapidly.
[0075] The preset parameters in the above formulas shall be set by those skilled in the art according to the actual situation.
[0076] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.
[0077] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.
[0078] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0079] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.
[0080] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0081] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0082] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A high-low voltage switch cabinet safety supervision intelligent optimization system based on end-cloud collaboration, characterized in that, It includes a terminal evidence processing module, a cloud evidence processing module, a proof obligation collaboration module, a danger time domain negotiation module, a safety disposal adjudication module, and a strategy modification module; The end-side evidence processing module is used to organize the switch cabinet field data into evidence fragments, and based on the evidence fragments, form a first risk statement carrying the end-side evidence domain that has been satisfied and the cloud evidence domain that has not been satisfied; The cloud-based evidence processing module is used to receive the first risk statement and evidence fragments, and after associating them with cloud context data, form a second risk statement carrying the evidence domain that has been satisfied in the cloud and the evidence domain that still needs to be verified on the end side. The proof obligation coordination module is used to determine the necessary evidence domain according to the risk type, generate proof obligations based on the evidence domains not satisfied in the first risk statement or the second risk statement, receive proof obligation feedback, and obtain proof support when the proof obligation involves multiple evidence fragments, so as to form a proof state by combining the closure of the proof obligation corresponding to the necessary evidence domain. The danger time domain negotiation module is used to convert risk characteristics into risk coordinates, calculate the danger time margin based on the risk coordinates, and generate the end-to-cloud division of labor results by combining the end-to-cloud collaborative time consumption. The safety handling decision module is used to determine the permissible action level based on the proof status, danger time margin, and local hard status, and to control the execution of on-site control actions according to the permissible action level and the local hard status verification results. The strategy correction module is used to correct the proof obligation parameters, danger time domain parameters, and action application conditions based on the event closure record after the risk event ends.
2. The high-low voltage switch cabinet safety supervision intelligent optimization system based on end-cloud collaboration according to claim 1, characterized in that, The edge evidence processing module is used to organize continuous sampled values in the field data into statistical segments and event-type data into event evidence segments. The statistical segments include the start and end times of the window, the mean within the window, the trend of change, and the peak value. The event evidence segments include the event trigger time and continuous data before, during, and after the trigger.
3. The high-low voltage switch cabinet safety supervision intelligent optimization system based on end-cloud collaboration according to claim 2, characterized in that, The end-side evidence processing module is also used to determine the credibility of the evidence fragment. The credibility of the evidence is determined based on the interval from the time of completion of the data collection to the time of participation in the judgment, the sensor state coefficient, and the local hard state consistency coefficient. The local hard state consistency coefficient is obtained by comparing the circuit breaker position, grounding switch state, cabinet door state, live display state, and lockout state at the corresponding time of the evidence fragment.
4. The high-low voltage switch cabinet safety supervision intelligent optimization system based on end-cloud collaboration according to claim 3, characterized in that, The first risk statement includes the risk object, risk type, risk location, triggering evidence fragment, evidence domain that has been satisfied, evidence domain that has not been satisfied, proof items to be verified by the cloud, and conditions for the statement to be valid. Among them, the content obtained on-site and meeting the credibility requirements is listed as evidence domain that has been satisfied, and the content that still needs to be verified in conjunction with historical operation curves, cabinet topology, operation and maintenance scenarios or comparison with similar equipment is listed as evidence domain that has not been satisfied.
5. The high-low voltage switch cabinet safety supervision intelligent optimization system based on end-cloud collaboration according to claim 4, characterized in that, The cloud-based evidence processing module is used to form a second risk statement based on the first risk statement and evidence fragments; the cloud-based context data includes at least one of historical operation curves, primary wiring topology of cabinets, equipment ledgers, maintenance work orders, repair records, and control action feedback records; the second risk statement includes evidence domains that have been satisfied in the cloud and evidence domains that still need to be verified at the end side.
6. The high-low voltage switch cabinet safety supervision intelligent optimization system based on end-cloud collaboration according to claim 5, characterized in that, The proof obligation coordination module is used to determine the necessary evidence domain according to the risk type. The necessary evidence domain is written in the form of a configuration file. The records in the configuration file include the risk type, necessary evidence domain, evidence domain attribution side, observable, triggering condition, judgment relationship, and whether the evidence domain belongs to the necessary evidence before control. When the necessary evidence before control is not closed, the safety disposal decision module does not upgrade the corresponding risk statement to an executable risk conclusion.
7. The high-low voltage switch cabinet safety supervision intelligent optimization system based on end-cloud collaboration according to claim 6, characterized in that, The proof obligation coordination module is also used to generate proof obligations for evidence domains that have not yet been satisfied, and to determine the proof obligation feedback based on the support, rebuttal, or indeterminate results of the feedback from the end side or cloud regarding the proof obligation; when the proof obligation involves multiple evidence fragments, the proof support degree is obtained by combining the support, rebuttal, or indeterminate results of each evidence fragment for the object of proof and the credibility of the evidence, and a proof status is formed for the security handling and adjudication module to call based on the proof obligation feedback, proof support degree, and whether the proof obligation corresponding to the necessary evidence domain is closed.
8. The high-low voltage switch cabinet safety supervision intelligent optimization system based on end-cloud collaboration according to claim 7, characterized in that, The danger time domain negotiation module is used to convert the risk characteristics corresponding to the risk statement and increasing with the degree of danger into risk coordinates, determine the danger time margin according to the rate of change of the risk coordinates, and calculate the end-to-cloud collaboration time based on the time reserved for communication round trip, cloud computing, proof feedback and control execution. When the danger time margin is insufficient to cover the end-to-cloud collaboration time, the end-to-cloud division of labor result for entering the end-side immediate handling process is generated. When the danger time margin allows the end-to-cloud round trip, the end-to-cloud division of labor result for entering the proof obligation supplementary proof process is generated.
9. The high-low voltage switch cabinet safety supervision intelligent optimization system based on end-cloud collaboration according to claim 8, characterized in that, The safety handling adjudication module is used to determine the permissible action level based on the proof status, dangerous time margin, and local hard status. The local hard status is used to determine the local electrical legality of the field control action. The field control action is executed after being verified by the end-side hard status. When the control suggestion issued by the cloud is higher than the permissible action level, the end-side will only execute actions that do not exceed the permissible action level and provide feedback on the reason for the downgrade. When the control suggestion conflicts with the local hard status, the end-side will refuse to execute it.
10. The high-low voltage switch cabinet safety supervision intelligent optimization system based on end-cloud collaboration according to claim 1, characterized in that, The strategy correction module is used to form a closed-loop record of the event after the risk event ends. The closed-loop record of the event includes risk statement, proof obligation, proof feedback, danger time margin, end-to-cloud collaboration time, action level, end-side execution result, retest result and manual confirmation result. The strategy correction module is also used to correct the proof obligation parameters, danger time domain parameters and action applicability conditions based on the retest result and manual confirmation result.
